THE MAKING OF RUNNING TIME
RUNNING TIME began
as an idea I had on December 31, 1995 and January 1, 1996, in other words,
New Year's Eve into the new year.
So I sat there and thought about it. Well, first of all ROPE is a play, all stuck in one apartment, so it's not great movie material to begin with (nor is it a very good play, either). Second, if one is going to go to the trouble of shooting in real time with no cuts, time must be the issue, which it isn't in ROPE. "Therefore," I continued to think,"what plot has an inherent time issue?"
A heist. I'll buy that for a dollar.
OK. Now, what are the technical problems? A roll of film is only ten minutes long, so, just like ROPE, there must be a hidden cut every ten minutes or so. Most of the "hidden cuts" in ROPE, however, aren't really hidden at all. In fact, several of the cuts are so awkward that I can easily imagine hearing the deleted soundtrack with Hitchcock's voice calling out in concern, "Jimmy, don't move. We're almost out of film and must now dolly into your back. Just freeze." Then the camera bouncily rolls into Jimmy Stewart's back to cover the hidden cut. So, where else can cuts be hidden? Going past dark objects (Hitchcock uses the lid of the trunk containing the dead body, which is OK, but it's not really moving). Going past pillars or other foreground objects is always effective, which Hitchcock never does in ROPE.
Then I really thought about it and realized that cuts can also be hidden in the blur of fast camera moves, commonly known as whip pans. Hitchcock didn't do any of those, either.
Then I thought, instead of just running each roll of film out to the very end and letting the hidden cuts fall where they may, it would be a much better idea to put the cuts in the best places possible -- where you'd naturally pass a foreground object or need a whip pan for story purposes -- somewhere around 7-9 minutes. Integrate the cuts into the story.
Also, if I wasn't going to shoot entirely on a set with controlled lighting, but was in fact going to go from inside to outside in the same shot, there would subsequently be color problems: different filters are needed for sunlight or incandescent light. A very simple solution presented itself rather quickly -- shoot in black and white. No color problems and it stylistcally fit the idea of a heist film; crime; noir, etc.
That was my revelation going into the new year of 1996.
For two weeks I mused, "A real time heist in black and white. Hmmmmm?" I came up with several entire plots that I didn't like and finally rejected.
I called my friend Peter and suggested that we kick the idea around. We got together and to my surprise and delight, instead of kicking it around, Peter pitched me most of the story. I was astounded; it was perfect -- exactly what I was looking for.
Now having a story, I spent the next three months writing the script. It was a very difficult script to write, too. After my first draft I never added another scene; all of the ongoing work was within the scenes that were there, expanding from within. The number of pages had to account for the number of minutes the film would run, give or take. It also had to be physically possible to do everything I was describing without a cut and a reset. There is a sequence when the truck has a flat tire, which is then fixed during the course of single take. It seemed like a cool idea, but could it be done? Could squibs (blood hits) be set off during the course of very long takes?
And since I intended to keep the camera moving as much as possible, where would the lights and boom go? And the crew, for that matter.
All good questions. Everytime I brought the idea of actually making this movie up to anyone they would display interest in the idea and complete disbelief that it could be pulled off.
I've done some pretty long takes before in other films -- one minute, two minutes, even three minutes -- but to do one very long take, two to three times longer than I had ever done before, after another after another did seem rather daunting. Nevertheless, I continued on . . .
I wrote the lead part with my good buddy, Bruce Campbell, in mind. If I could actually convince Bruce that I was making this movie -- considering that I had made my last film seven years earlier -- then I would be a long way toward honestly making it.
Well, I somehow not only convinced Bruce to star in the film, but to invest in it as well (now that's my kind of actor).
Then I offered the second-lead to Jeremy Roberts, an actor I had directed on XENA: WARRIOR PRINCESS. Jeremy read the script and accepted. My two leads were cast.
I ran into an actress that I knew slightly from Detroit, Anita Barone, at a party. I'd seen her in several plays and knew that she was a first-rate actor. Without thinking twice I offered her the part. She subsequently read the script and accepted. Now I had my entire lead cast, and they were all my first choices.
Since Bruce, Jeremy, and Anita work all the time, the next thing I had to do was choose shooting dates that would accomodate all three of their schedules. I chose my birthday, August 17, five months hence. Bruce informed me upon my declaration of the "locked-down, absolutely no chance of changing" shoot dates, that if I didn't shoot the film then, he'd kill me. I finally ended up pushing the date back two weeks. Luckily, Bruce didn't kill me.
My friend Jane had just broken up with her mate of 12 years and had also quit her job of 10 years with Steven Spielberg. I asked her if she'd like to produce the film with me and she agreed. I would entirely handle the legal and monetary end of things, she and I together would handle the actual producing.
I must digress again. My optimism about having the money to make the movie was based primarily on my recent sale of a script to the company that produced AIR FORCE ONE, thus giving me some disposable income to squander. I was hoping to put up about half the money myself and raise the other half from friends and loved ones. I ended up putting up more like three-quarters. The final budget was $120,000.
In March of 1996, Jane and I began scouting locations. Although the story is very limited in its scope, it still has numerous locations, many of which that had to co-exist directly next to other locations because we couldn't cut between them. In the succeeding months, Jane and I drove up and down every side-street and alley in Santa Monica, Venice, West L.A., West Hollywood, Burbank, Northridge, Van Nuys, Glendale, and downtown L.A. We looked at warehouses and abandoned factories all over the city. Our locations finally ended up being: downtown L.A., Glendale, Santa Monica, and the prison location which was in Lancaster, 90 minutes north of L.A.
After Jane, the next person to jump aboard was the casting director, Donise. She put out the call for all of our other actors on the internet and within a week she had over a five foot pile of headshots, literally 1000 of them. We culled it down and culled it down until we auditioned 60 actors for five parts.
I then made two important decisions that I have not fully contended with even now, a year later. I chose 16mm, Kodak, ASA 64, black and white stock, and I decided to shoot it full- frame at a screen ratio of 1:1.33.
First of all, ASA 64 is a relatively slow-speed film stock, meaning it's thicker and has more chemicals on it, but needs more light to penetrate those chemicals and expose. The result of using slow film stock is finer grain and a sharper image. I personally just like the look of a fine grain image, which isn't used very much anymore in movies even in color, but hardly at all in black and white (I had to special order the stock). Beyond that, if I ever needed to do a 35mm blow-up -- which I'm just about to do -- the image quality should hold up better. We'll see on that very soon.
The second decision was choosing to shoot at 1:1.33. The average movie right now is shot at 1:1.85, meaning the picture is 1.85 times as wide as it is tall, almost 2 to 1. At 1:1.33 the picture is nearly square; very similar to the size of a TV screen. With only a few experimental exceptions, all movies were shot at 1:1.33 from the beginning of motion pictures until 1953 when Cinemascope was introduced. That's why all old black and white movies from the 30's and 40's look fine on TV. Until the advent of letter-boxing, all the wide-screen movies of the 50's and 60's just look wrong on TV with half the image cut off. The average movie shot at 1:1.85 is losing more than 25% of the image when shown on TV (unless letter-boxed).
One popular method now of bypassing this problem is to shoot Super-16mm, which has a screen ratio of just about 1:1.85, thus blowing-up almost perfectly to the present 35mm theatrical standards. The way that Super-16 achieves this wide-screen format is that instead of running double-perf film stock, meaning there are sprocket holes on both sides of the film, it runs single-perf, which is in fact print stock. 16mm print stock only has perforations on one side so that the optical soundtrack can go on the other side. Since there is no optical soundtrack while shooting (sound is recorded on a sound recorder and transferred to optical in post), that allows extra space for the image to cover, thus giving you the 1:1.85 screen ratio. Here's the big problem: you can't make a 16mm print -- you've used up the area where the soundtrack goes -- so now you must blow-up to 35mm to show the film theatrically. Since 16mm prints are less than half as much money as 35mm prints, not to mention the cost of the blow-up, I decided that I wanted to make 16mm prints to send out to the film festivals. Also, Super-16 camera equipment is considerably more expensive to rent than regular 16mm. Beyond that, aesthetically, wide-screen didn't seem appropriate for the subject matter of the film nor the technical problems that I knew I would be facing trying to hide light stands, booms, and the crew on film with such long takes. So I chose 1:1.33. Hey! CITIZEN KANE looks terrific at 1:1.33.
Jane and I then put together our small, young crew.
I have been a 2nd unit director on TV movies, and my 14-person 2nd unit crew could do anything the 50-person main unit crew could do, but faster. Having this invaluable knowledge, I now intended to put together very small, lean crew.
On RUNNING TIME the entire camera/lighting department consisted of: the Director of Photography, the Steadi-Cam and camera operator (the same person), the 1st Assistant Camerman, a key grip/gaffer, and a loader, and that was the biggest department by far.
There were no assistant directors, however the production coordinator took over the position on set by sheer attrition (I was usually stationed in another room with the wireless monitor). Bruce Campbell also did his bit as 1st A.D. He'd regularly get fed up with what he considered extra dicking around and suddenly do a perfect impersonation of an impatient 1st A.D., clapping his hands, "All right, that's it! We're done! Off the set! Now! Move it!" And before I knew it we were shooting, which was lovely.
We shot for ten days -- two five-day weeks -- and wrapped early every single day. On the sixth day of shooting we got so far ahead of schedule (we shot 14 pages instead of just 8), that on the seventh day we ended up wrapping at 10:30 A.M. because I couldn't reschedule the effects guys to come earlier. Jane and I ate our catered lunch for 20 by ourselves.
(As an amusing sidelight, a mutual friend of Bruce's and mine from Detroit, John Cameron, was working as 1st A.D. on MEN IN BLACK simultaneous to our shoot. Each day he and Bruce would speak on their cell phones and compare notes about how much got shot -- RUNNING TIME: 8 pages; MEN IN BLACK: an 1/8 of a page; RUNNING TIME: 14 pages; MEN IN BLACK: a 1/4 of a page, etc.).
A big portion of act one takes place in the afore-mentioned truck. The way this sort of sequence is generally shot is either: (A) pure fakery, meaning people are shaking the truck to make it look like it's moving while other people are swivelling lights past the truck to give the impression of movement, or (B) using rear-screen or blue screen projection, having a section of the truck set up on a soundstage and projecting a moving image of the passing road behind the actors, or C. towing the truck behind another truck that is also towing a generator to power the lights. For the sense of realism and practicality, I wanted to shoot in an actual moving truck
driving through city streets. Since we could not afford a real, quiet, movie generator, or another truck big enough to haul our picture truck, the answer, in my mind, was to create a portable lighting grid in the truck powered by DC current from a car battery. Also, since our space was very limited inside the truck, there was a reasonable chance that whatever lights I used would be seen, so they had to be of a nature that might actually be in a truck. I purchased ten black, clamp lights, tore out their guts and replaced them with low-wattage, DC, movie-projector lamps and bulbs, then wired them all into an extra-large car battery. These were all clamped to strips of wood attached to the truck's ceiling, all of which could be slid back and forth or side to side. It all worked brilliantly, I'm both happy and proud to say.
Squibs are usually wired into an actor's costume, the wires or "tail" running out their pant leg, then detonated by a pyrotechnician sitting out of frame. This would not be practical with long, extended takes. I asked my good buddy, Gary Jones, who had done special effects on my first two films, how I should handle the squibs? Gary's sharp response was to have the actors set off their own squibs -- the detonator switch running down their shirt sleeve into their hand. Since there would be blanks in the guns, when the actor heard the blank fire, they'd push their button -- Pow! -- and die. Once again, it worked great.
Shooting the very long takes quickly turned out to be difficult, but a real time-saver. Once you get it, you've got it; there's no close-ups, everse-shots or inserts to pick up. What we also realized pretty fast was that after two or three rehearsals, even if it hadn't come together, which it generally hadn't yet, it was time to shoot it. Basically, you don't get a 100% from anyone -- nor do you even want it -- until you run film through the camera.
Actors remembering their lines was not a problem. We had four rehearsals of a couple of hours each where we worked out a lot of blocking and line-changes. The actors knew which 7- 10 page hunks we'd be shooting everyday, and 7-8 pages is what any actor in a leading part will have to learn for any given day of shooting on a TV show anyway, so it was no big deal. Plus, of course, I used good SAG actors, the single largest cost on the film (one-third of the budget), and worth every cent. You can't put your money in a better place on a low-budget movie, in my opinion. I'd rather have no effects and good actors then a lot of effects (be they good or bad) and bad actors. But that's just me.
For the first time in my life on RUNNING TIME I shot entirely in order, moving from scene one to scene two to scene three to the end. It was great. Showing up to work everyday was like going to see a serial. When we shot the last scene we were done and the movie was over. It caused some scheduling problems, but was well worth it, particularly when shooting this type of film, in real time, one thing leading directly in to the next.
In most things, even if there's a line-flub or the boom drops in or a light flares or whatever, you just keep going, some part of it is probably usable and you'll cut in and out to get it. When shooting in long takes, if something goes wrong, it's all wrong and you can just stop. Our biggest ongoing problem was at some point or other catching the poor boom man. Still, this was far preferable to using wireless microphones (which we did revert to a couple of times), just because the boomed mike sounds better. This also allowed me to use a few oldtime Hollywood lines that amuse me. When, six minutes into a take, suddenly there is the boom man looking sheepishly into the lens, pretending to suddenly be invisible, I would yell: "If you join SAG you'll get residuals." But our boom man was excellent and young and it was not a big problem.
I did another odd thing that worked out wonderfully well. At the end of every shot all the actors would group together under the microphone and run the entire sequence again, without a camera motor buzzing or clothes rustling or footsteps or anything else. This allowed me to eliminate thousands of dollars of voice-replacement in post-production. I did do one hour of looping with Bruce adding some huffing and puffing to all of his running. That's it.
Our ten day shoot, wrapping early every single day, went by so fast it's like it happened in a dream. There were various other moviemaking shenanigans with police and vehicle permits and a creep that rented us a stage, then made us pull our set down, then rebuild it the next day. Movies are just like that and I love that about them; as hard as you try to plan and scout and prepare, there's always the unexpected which you must contend with as well.
I purchased 15,000 feet of film and ended up shooting just over 10,000 feet. The final film in 16mm is 2800 feet long, so my shooting ratio was 3 1/2 to 1. I sold my unused film to Steadi-Systems (which, like Studio Film & Tape, buys unused film stock and short-ends). I then decided to shoot two more scenes, set up my reshoot and had to buy my own damn film back from Steadi-Systems at $20 extra a roll. The moral of this story is: don't get rid of your extra film until you're absolutely sure that you're done shooting.
Every exterior shot had to be planned with the time of day and the position of the sun in mind so as to avoid camera shadows -- and even still I got a couple.
To keep the camera mobile on the interiors I decided to circle the actors on many occasions. To achieve this an entire conga line is always following the camera. In the girl's apartment at the end of the film (which is in fact my apartment), to get the camera fully around the actor's sitting on the bed (which I did several times), I first had to make sure that the headboard was sturdy enough to support a guy wearing a Steadi-Cam and an Arriflex SR camera. Next, for the cameraman to step up onto the headboard a person had to set a box under his foot as a step, then about ten stuffed animals had to be cleared away from around the cameraman's feet, then another box was put in so he could step down from the headboard, then all of the stuffed animals were replaced so that as the camera continued around everything looked normal. Quite frankly, it's all unnoticeable in the film itself, which is as it ought to be.
There are 30 cuts in the film, 32 with the front and end titles attached. Since, unlike any other movie but ROPE, no more cuts could be made under any circumstances, the editor and I spent a lot of time experimenting with those 30 cuts, a frame this way, a frame that way, two frames this way, two frames that way, etc. I'd say 20 of the 30 cuts are perfectly hidden. 10 of the cuts are dicey.
Having now worked in this style of filmmaking I would not only happily do it again, but I'd do a much better job the next time. I got caught in several of the same traps as did Mr. Hitchcock -- movement straight into the lens makes for a clumsy cut; movement across the lens makes a much better cut.
RUNNING TIME was cut on a non-linear editing system, the D-Vision, which is already almost obsolete. I have cut on both Avid and Lightworks, the other non-linear editing systems, and they're all great. It is definitely the way to go if you can afford it. In my case, my friend, Kaye, owns this D-Vision machine, and not only edited the film for free, but threw in the use of her equipment, too.
Also, my good friend, Joe LoDuca, who composed the scores for my first two films (as well as all three EVIL DEAD films, and HERCULES and XENA), composed and recorded the score for RUNNING TIME for free.
I think both Joe and Kaye were so impressed that I actually got another movie made after so many years that they helped me out of the sheer kindness of their hearts. I gratefully thank them, too, as well as everyone else that worked on the film.
The post-production sound was done by Ideal Sound, a small company that specializes in low-budget features (many of the films they've done are parts two, three, and four of films you've never seen part one of). These guys were completely wonderful. For one flat (and affordable) price they did everything (I won't say how much because everyone has to make their own deals). They cut all of the sound effects, background tracks, and dialog, then mixed the film very well in both stereo and mono (stereo for the video and 35mm tracks, mono for the 16mm track). Our 16mm prints were made by Hollywood Film Lab, a small and very friendly lab right in Hollywood.
So when can you legitimately call a film done? When you make release prints? When the film is actually released, which in many cases is never. Or perhaps when the bills stop coming in. In that case RUNNING TIME is not done yet.