Skip to main content

Format Face-Off: Film vs. Digital Video

By David Heuring

A scene from Eye-Trap, shot by Thomas Repp on ARRI 435 and Sony HD F-900 24P.

Like most filmmakers, Thomas Repp, BVK, has heard the rhetoric comparing the pros and cons of originating on film and the new HD 24P video system. In an effort to find out for himself, Repp decided to make a short film comparing an ARRI 435 film camera and a Sony HD F-900 24P camera in real world situations.

“I never had a problem switching between film and video,” says Repp, who has compiled an impressive body of documentary work on German and French TV. “I’ve worked in just about every format, so I was enthusiastic when I heard about the 24P HD camera.”

Repp was interested in technical comparisons, but he also wanted to see how the choice of media affects the emotional impact of the stories. “I believe that a main difference between film and video is how they record surfaces (or textures), especially when you are magnifying big images on movie screens,” he says. “I wanted to see if this new 24P digital camera makes a difference.”

Step one in the process the story: Die Augenfalle (Eye-Trap) is an original short story written and directed by Elena Alverez. It was inspired by photographer Jacques-Henri Lartigue, who was born in 1901 and recorded more than 200,000 photographs.

Thomas, the boy at the center of the story, is a shy teenager who captures images he sees by blinking his eyes. He saves the images in his mind’s “eye-trap.” Thomas becomes enchanted with a beautiful teenage girl with a red silk scarf.

Repp worked a dozen or so situations into the script. “I wanted to test how surfaces come across and I also wanted some shots in low-key light with silhouettes,” he says. “I also wanted to see skin tones in beauty shots with an actress in high key light. I wanted handheld shots with fast pans to see how the compression in the 24P camera handles that situation compared to film. That’s important, because when you are doing a fast pan there are differences in the image structure from frame to frame. I also wanted to compare differences in handling the cameras and the use of filters.”

Eye-Trap was produced at practical locations in Munich, Germany. Repp used prime lenses on the ARRI 435 camera with Kodak Vision 250D film for daylight exteriors and Kodak Vision 200T films for interiors and night scenes. He used Fujinon lenses on the Sony HD F-900 camera. Repp shot each scene first with the HD camera and then with the film camera working with essentially the same crew.

They made about 80 shots in two days. Support was also provided by ARRI-Cine Technik, ARRI Digital, Media AG and Kodak AG. Repp also wanted to gauge how each system handled vibrations and whether there was vertical smear in the 24P camera.

“We shot one scene from a car traveling 70 miles an hour on a bumpy road, so we could see how the vibrations affected the contact between the recording head and tape,” he says.

There are also scenes with extreme ranges of contrast. Repp wanted to evaluate differences in image quality. One shot included a group of boys playing in a sunlit meadow along with Thomas, who is standing beneath a shade tree, writing in his booklet.

“There is an extreme range of contrast between the sun-lit and shaded elements of this shot,” says Repp. “You can see the details in the highlights on the film. The highlights blow out on video. There is another shot where the boy is in a room lit by a lamp. He switches the light off. I wanted to see how far we could see into the darkness. He was wearing a yellow T-shirt because I’ve had problems with other video cameras ‘seeing’ yellow.”

Repp was also interested in how the two media rendered colors. “We outfitted the girl in one scene in a black coat and an unbleached white scarf, because I wanted to see the difference,” he says. “We got the same results every time. The film camera recorded a richer image.”

He also compared latitude. “I noticed there is much less latitude with the 24P camera than with film,” Repp says. “Film records many more tones and colors. The digital video camera lost details, particularly in the highlights, and you also don’t get the same feeling of separation (between the foreground and background).”

Repp notes that he was very happy with the video rushes on HD monitors, but disappointed when he saw the same digital images projected onto the large screen. “The focus wasn’t as sharp as it had appeared on the high-def monitor,” he says. “We actually decided to re-shoot some scenes. I think that the problem is that video rental houses aren’t quite as critical about testing lenses with different bodies as film houses. Unless you have a tight tolerance between the lens and the camera body, there will be some drift in focus.”

A take of a red scarf against a blue sky was shot at 100 frames per second with a polarizing filter on the ARRI 435. “We stretched the shot out for dramatic purposes,” says Repp. “I’ve always been impressed by the fact that during the days of silent films, cameramen could crank film at different speeds to enhance certain moments in the story-telling. It's difficult to imagine shooting a film today without slow motion. It is becoming part of our visual language like it was in silent movie days.” He also used faster frame rates to smooth out certain shots.

Repp concluded that over- and undercranking is impossible to adequately duplicate using video techniques. There is an electronic post process where images can be sped up or slowed down, but he says it looks “fake.” “You don’t have that poetic, fluent movement—it shutters. In scenes like the one where the boy turns off the lamp, the HD camera did a surprisingly good job seeing into shadows, however, it didn't handle highlights nearly as well.

“Details burned out and colors were desaturated,” he says. “Sometimes you get these ugly, big spots that are too bright and distracting. The highlights blew out if they were two stops over-exposed, and at two-and-a-half stops all the details were gone. That makes high key shots very critical because you operate in a range of two-to-four stops over-exposed. With film, you still have details when you are four stops over-exposed.”

Nevertheless, Repp says that the HD cameras are a significant advance for the right applications. “It’s an improvement for projects that formerly might have been done in DigiBeta format,” he says.

“Think of the possibilities in Dogme-type films, where you have high shooting ratios that are not economical for film. It is the same for shooting at live events, where you have a lot of coverage for a long time. I don’t see it as a replacement for film because it takes away too many possibilities—the ability to use different frame rates, a freer choice of formats and the freedom to use more dramatic lighting. I think film also gives you an ability to work faster on exterior shots because you aren’t relying on a high-def monitor.”

Repp notes that the new HD camera has the same size chip (two-thirds of an inch) as the digital video. That is an area much smaller than a 35mm frame. “That means you need a wider-angle lens that gives you more apparent depth of field, to the point where there is no possibility of separation between the background and foreground,” he says. “They seem like they are glued together without any feeling that it's three-dimensional, even in static shots. In beauty shots, you lose all that feeling of dimension.

“At the end of the film, when the boy has photographed the girl, there are lights and branches in the background,” he explains. “In the film version, the lights blur, creating a heavenly atmosphere. It’s the climax of the story—the pay-off. It is a simple, but very effective technique. The 24P shot has a much more realistic look. It shows facts instead of creating this fantasy atmosphere.”

Based on his experience, Repp says, “There is more than one type of documentary. Some of them are kind of peep shows for the audience. It’s right for those programs to look like reality. My type of documentary is usually more abstract, and you want the audience to know there is some interpretation.”

Regarding the Eye-Trap test, Repp says, “I can only speak for myself and tell you what Elena and others who worked with us have observed. After comparing the results, we chose to send prints made from the 35mm film on the festival circuit. But we are showing the film and HD versions at these screenings so people can decide for themselves.”


David Heuring is a writer at Creative Communications Services. He is the former editor of American Cinematographer magazine.