Directing the Camera

by Gil Bettman

  Board #15 — Figure 6.004

  Board #15 is an insert. An insert is a tight shot that usually reveals some small crucial detail or bit of action. In almost all inserts you do not see an actor’s face. Instead you focus on some part of his body — a wound, for example — or what he is doing with his hands, like pulling the pin of a grenade. When shooting inserts there are two important criteria that must be met:

  ● They should be easy to read, which is to say, the audience should be able to immediately understand what they are looking at.

  ● They should cut smoothly into the overall action.

  The camera setup in Board #17 is drawn to meet these two criteria. The shot is tight enough so the audience will be able to see at a glance that Ragnar is reaching for a crowbar hanging from a bracket on the side of the elevator shaft. Ragnar will initiate the motion of reaching in the preceding Board #14. This reaching motion will be continued and completed in Board #15. By cutting on this action the editor will be able to make a smooth, seamless edit.

  The significance of Board #15 and its importance in the overall sequence cannot be over-emphasized. This is where the ambush actually starts. The insert in Board #15 instantly and effectively makes the danger real and imminent. For this reason, in the final sound mix of the film, the ominous music signaling the instant escalation of the threat level started with this shot.

  And yet, of all the shots depicted thus far in this action sequence, the insert in Board #15 was, by far, the fastest and the easiest to do. Because inserts are tight shots on small, simple motions, like reaching for a crowbar, they can be quickly framed up, rehearsed, and shot. They usually require little or no lighting. They are cheap pieces — the cheapest you can get. Therefore:

  ● When you are doing an action sequence on a budget, you should shoot every possible insert you can think of.

  Because, as I explained in Chapter 5, when it comes to action, the more pieces you have, the more energy you can pump into the action in editorial. Shooting every insert you might use will help you get the maximum bang for every buck you have in the budget.

  So, if you are shooting a car chase, you should shoot an insert of everything the driver does to drive the car: brake, accelerate, shift, and steer. You can also do a tight shot on the speedometer. A camera operator, working with one assistant and an extra, can do all of these shots on a soundstage in about an hour and a half. That comes out to twenty minutes per shot. The shot in Board #15 probably took no more than twenty minutes to knock off. And yet it generates almost as much suspense and has as much importance in this ambush sequence as Board #23, which probably took at least ninety minutes to set up and execute, because it involves a semi-dangerous stunt that was shot with multiple cameras. Inserts can be as powerful as any shot, and yet, they take about one-fifth as much time to do. So, shooting all the inserts you might use is a very wise strategy.

  FIGURE 6.004

  Board #16 and Board #17 — Figure 6.004

  The purpose of Board #16 and Board #17 is to telegraph to the audience that Lance is, unsuspectingly, falling into the trap set by Ragnar. The audience knows there is danger and knows where it is, but the hero does not, and the hero is unwittingly moving closer and closer to the danger. Such moments use the narrative device of dramatic irony to jack up the level of suspense in a film. Dramatic irony works by making sure the audience knows more than the characters in a film. It taps into our natural, human response to want to intervene on the side of those we identify with when they are facing peril. So the story is best told by prolonging this moment of suspense for as long as possible. Board #16 and Board #17 would ideally serve this purpose.

  I drew Board #16 to be shot with an extreme wide-angle lens directly behind Lance and Board #17 to be shot with an extreme wide-angle looking up at Lance through the handlebars of his motorcycle, because I knew I could put a small, lightweight 35mm camera, known as an Imo, in these positions. When the stunt coordinator and I had visited the dam and laid out this sequence, he had told me that our key grip owned the camera mounts with which we could mount an Imo camera in the positions depicted in Board #16 and Board #17. In these two instances, the look of the shot was largely determined by where the mounts could be attached to the motorcycle. Practicality also dictated the lens selection. It was absolutely necessary to use an extreme wide-angle lens, because the cameras mounted in these positions would be subjected to a great deal of vibration. The wide-angle pushes objects in the frame into the distance by making them smaller. So it dampens down vibration because the up-and-down motion of the lens is less noticeable against the smaller, more distant objects in the frame.

  Mount shots are a great tool for the low-budget action director. With the small expense needed to double vehicles and mount cameras he can add a monumental amount of production value and visual energy to a chase sequence.

  I also drew the shot in Board #17 because I knew John Stamos could actually drive the motorcycle with the Imo mounted on the front fork. This made this setup an extremely valuable shot, which I knew I had to get to make this action sequence come to life.

  When you are shooting an action sequence, you must make a concerted effort to shoot tight shots on the “real-ies,” the actual actor, in this case, John Stamos, going through the same motions as the stuntmen.

  The stunt sequence is only exciting if the audience truly fears that the good guy(s) is really in danger of being defeated or even killed by the bad guy(s). The best way to make this danger real is to keep showing the actual faces of the actors in close-up going through the motions required to do whatever is being done in the stunts. In a stunt sequence, as in a dramatic sequence, the center of the drama is in the eyes — in particular the eyes of the real-ies. The shot that shows the most of the eyes is the frontal close-up — the closer the better. When shooting an action sequence, it is easy to get carried away shooting wide shots with the stunt doubles because these shots convey the power of the stunts. But this is not enough. You have to keep shooting tight shots of the real actors. This keeps the action an integral part of the story of the film and prevents it from becoming a dehumanized spectacle, the outcome of which the audience cares very little about. Therefore it is absolutely necessary to:

  ● Put the real-ies in the action as much as time and money allow.

  The importance of this cannot be over-emphasized. The setup in Board #17 gave me a good frontal close-up on Stamos and did an excellent job of conveying what his character, Lance, is feeling as he hunts for Ragnar, but in so doing, falls into a trap.

  If you require an absolute confirmation of the importance of shooting close-ups on the real-ies, rent a DVD of your favorite action film and go through the stunt sequences with an eye for how many times you see the actual actors’ faces throughout the sequences. You will find that the shots on the real-ies are frequent and continuous.

  The shot in Board #17 — Figure 6.004 — is also an extremely valuable shot because it is a neutral shot. It is neutral because the only background visible is the blue sky above Lance’s head. The look of the sky behind Lance is not going to change if Lance is five feet or five miles away from Ragnar. So, this shot can be dropped in editorially anywhere in the sequence and it will fit. This makes it valuable for two reasons. First, you can cut to it repeatedly and use it to prolong the suspense. And, second, you use it as an editorial bridge to cut around the weaker moments in the action.

  The value of a neutral shot such as the one depicted in Board #17 — Figure 6.004 — is borne out by the fact that, even though in this pre-visualized storyboard version of this ambush I anticipated cutting to it twice, in the final edited version of the film it was used five times.

  Board #18 — Figure 6.004

  The shot in Board #18 is the same as that depicted in Board #13 and it was intended to be used in the same way — to generate suspense by showing Lance proceeding gallantly ahead, unaware that he is on the verge of being ambushed by Ragnar. As explained above, through the use of dramatic
irony, this would draw the audience deeper into the story and make them actually fear for Lance’s safety.

  This shot would best be done with a 50mm lens or longer. The use of the long lens would slow Lance’s motion coming toward the camera and make him seem to hang suspended in the frame. This would prolong the moment of truth and add to the suspense.

  Board #19 — Figure 6.004

  Board #19 depicts the moment of truth. Ragnar strikes by driving the crowbar into the spokes of Lance’s motorcycle. Without a doubt, this is the best angle for this shot, because it makes the violence and danger seem real, whereas in fact, they were faked. From this angle behind Ragnar, the stuntman doubling him can make a powerful thrust forward with the crowbar and it will seem to go into the spokes of the motorcycle, but the stuntman will actually pull up short of the spokes.

  Board #20 — Figure 6.004

  This board depicts an insert of the crowbar going into the spokes of the motorcycle. The shot illustrates another key way that inserts are useful in action sequences. Inserts can enable you as a director to fill in a missing piece in the action by using a tight, controlled environment to shoot something that would be dangerous or impossible to do in a live action stunt.

  It was absolutely necessary to shoot this tight shot of a rubber crowbar being thrust into the spokes of the rear wheel of Lance’s motorcycle. If we had only done the shot in Board #19, it would not have been believable that Ragnar had actually done this. The danger would not have seemed real. The suspense would have been diminished.

  Board #20 is a three-quarter angle from slightly above the rear wheel of the motorcycle. The three-quarter angle gives depth to the shot and reveals that the rubber crowbar actually went between the spokes. The tight frame and downward angle conceal the fact that nobody was driving the motorcycle, but instead it was being pulled through the frame by a pair of grips.

  Board #21 — Figure 6.005

  This is the same shot I drew in Board #17. I included it as a visual reminder to follow the rule stated above and:

  ● Put the real-ies in the action as much as time and money allow.

  In this shot, Lance is depicted losing control of the motorcycle. To make the audience believe that this was actually happening, I needed to be in a tight shot on the real-ie — in this case John Stamos — in order to show the look on his face, and, in particular, the fear and confusion in his eyes.

  To get the shot, I instructed Stamos when he actually drove the motorcycle with the Imo camera mounted on the front fork to throw his body from one side to the other to simulate the motion of losing control. When we did the shot, because Stamos was in complete control of the bike, his simulation actually looked quite phony (see Figure 6.008 to 6.012).

  This is usually the case with such simulated shots on the real-ies. For this reason, the actors, who tend to be insecure by nature, will sometimes refuse to “go through the motions” in this way. What they are actually afraid of is looking silly in front of the macho stuntmen and crew. Rather than admit this, they will rationalize by refusing to do the shot, because “it will never match” what the stuntmen do. You must patiently and firmly refuse to let them get away with this. These tight shots on the faces of the actors at these moments of danger are absolutely crucial to the success of the finished piece, and you must fight for every one of them.

  Boards #22, 23, and 24 — Figure 6.005

  These three boards were drawn to depict how I planned to shoot the actual stunt of the stuntman “laying down” the motorcycle. I anticipated that I would be able to shoot this stunt simultaneously with three cameras. The shots in Board #22 and #23 would be done with an Arriflex and operated, whereas the shot in Board #24 would be done with a locked off Imo in a crash housing.

  Such a plan to shoot a stunt with a couple of operated cameras and any number of locked-off Imos in crash housings is the most reliable and efficient approach. In particular, when doing car or motorcycle stunts, if it involves any real danger, you want to minimize the actual number of times the stuntman or stuntmen have to drive the stunt. Therefore, you shoot it with as many cameras as you reasonably can. This saves time and money. Every time a stuntman does something dangerous he is paid an adjustment, which is a specific dollar amount on top of his day rate. Whoever is writing the checks will expect you as the director to keep down the number of adjustments. Equally important, every time danger is involved, it takes time to set up the shot. The stuntman should not be rushed. And the cameramen need to make absolutely sure they get a good shot. Generally speaking, when you are working on a low-budget production, if you have to drive the stunt twice, then you have screwed up. Your goal should be to do it well the first time and move on.

  Shooting low budget, you rarely have enough money to shoot with more than two operated cameras. Good operators are expensive. This tends to limit their number at two.

  6.008

  6.009

  6.010

  6.011

  6.12

  FIGURE 6.005

  The assistant cameramen, in addition to loading and unloading all the cameras, do most of the work of setting up the unoperated, “locked-off” Imos in crash housings. Added Imos require added assistant cameramen. This tends to limit the number of Imos that can be used effectively at three.

  To shoot a shot with multiple cameras you have to make sure the cameras do not see each other. If they do see each other, it should be at moments that can be easily edited out. In the case of Boards #22, 23, and 24, the most efficient way to decide where to put the three cameras would be to start by determining the location of the one camera that could not be moved — the camera that would shoot the shot in Board #23, and then adjust the other cameras accordingly.

  The reason the camera in Board #23 could not be moved was because it was at the top of the staircase that ran up outside the elevator shaft and led to the machine room at the top of the shaft. Doug Lefler drew this exterior staircase in Board #16. An operated camera shooting down from this spot, twelve feet above the top of the dam, would get a great shot of the bike being laid down, because it could show the entire length of the stunt — the bike going down and the stuntman sliding along on his back behind the motorcycle. In this case, the motion would be across the frame, or more accurately, from the bottom of the frame to the top, so the motion would be accentuated and the shot energized if it was shot with a long lens. However, as with Board #5 above, the long lens would make the bottom of the dam seem closer and therefore diminish the depth of the drop. It was important to see the depth of the drop in order to show the audience that Lance, having been knocked off his bike, was in danger of falling to his death. The depth would seem greatest with a wide-angle lens. So I split the difference between the two lenses and drew this setup as if it were to be shot with a “normal” — 50mm — lens.

  Since the shot in Board #23 could only be done in one spot, the two other cameras used to simultaneously shoot the bike lay-down had to be positioned so the three cameras would not see each other. To this end, the cameraman who would shoot the shot drawn in Board #22 would make certain that his camera was just out of frame at the bottom of the shot depicted in Board #23. Similarly, the camera that would shoot the shot in Board #24 would be carefully positioned so that it would still be out of frame at the top of the shot depicted in Board #23.

  Board #22 would be shot with an operated camera, since it would require a slight right-to-left pan. To accentuate the motion of the motorcycle, the camera would be placed at the same height as the axles of the motorcycle, and the stuntman would try to drive as close to the camera as he could without hitting it.

  Board #24 would have to be shot with an Imo in a crash housing, because of the very real chance that the bike could hit the camera. The lens used would be a wide-angle to accentuate the right-to-left motion of the bike and rider as they slid by the camera.

  Board #25 — Figure 6.005

  This board is the first of the series that shows the motorcycle going over the edge of t
he dam and falling the full height of the dam, before smashing to pieces on the rocks below. This series of shots graphically establishes the full extent of the danger now facing the hero, Lance. If the motorcycle goes off the edge and smashes to bits on the rocks below, then Lance is next. Only a small miracle or a last-minute circus-act catch is going to prevent him from falling to his death. If shot properly, this should produce a moment of intense suspense for the audience and add to the excitement of the film.

  Unlike all the other shots in this sequence, when I drew Board #25 with Doug Lefler, I had not previously planned out with the stunt coordinator and the DP how I was going to do the shot. So there is no lesson to be extracted from this board. It was only drawn as a placeholder to signify that we needed to shoot this piece of the action.

  Figures 6.013 to 6.015 are frame grabs from the shot we actually did. To get this shot we positioned the camera so the lens was about two feet in the air, straight above the very tip of the arrow drawn in Board #25, pointing down so the camera could see over the lip to the drop below. This sold the height and the danger of the fall. Then the grips laid the bike on its side and shoved it out over the edge under the lens of the camera, without dropping it. We used a 50mm lens. A wider lens would have made the drop seem greater, but the expanded field of vision of the lens would have also revealed the grips holding on to the motorcycle, just out of frame, camera left.

  6.013

  6.014

  6.015

  Board #26 and Board #27 — Figure 6.005 and 6.006

  These two boards were drawn to show the motorcycle falling from a great height. As it turned out, we only did the shot in Board #27. It accomplished the desired result of showing the motorcycle falling, and it was by far the easier shot to do, because it could be shot from the top of the dam. Whereas getting the shot in Board #26 would have required the additional time needed to send a camera assistant down the elevator to the bottom of the dam in order to set up an Imo in a crash housing.