Today we’re discussing The Call of the Wild which, like the recent Planet of the Apes movies, is a blend of live-action photography and CGI. The two editors on the film are William Hoy, ACE, and David Heinz.
William Hoy, ACE got into the editor’s seat on feature films back in the mid-‘80s. One of his first major feature films was Dances With Wolves. Since then he’s edited a string of box office and critically acclaimed hit films: Star Trek VI: The Undiscovered Country, Patriot Games, Se7en, The Man in the Iron Mask, The Bone Collector, Dawn of the Dead, Fantastic Four, 300, Watchmen, and Dawn of the Planet of the Apes.
Editor David Heinz previously worked with William as the VFX editor on both War for the Planet of the Apes and Dawn of the Planet of the Apes. He also held that position on The Jungle Book. His other editing credits include numerous feature films and TV shows.
David actually started on The Call of the Wild an entire year before production began – cutting previs.
This interview – and many others – are available as a podcast. (This interview is available Tuesday, February 25th)
(This interview was transcribed with SpeedScriber. Thanks to Martin Baker at Digital Heaven)
HULLFISH: What are some of the difficulties of dealing with characters that are so important to the story that are CGI?
HOY: In comparison to a picture like Apes for instance — the two Ape movies — we had motion capture actors that were acting emotionally and I actually edited those motion-capture performances (as stand-ins for the finished CGI.)
I noticed that a lot of reviews are crediting Terry Notary for being Buck, the dog. Not to take anything away from Terry — he’s a great motion capture actor — but in actuality, we didn’t use very much of the actions other than the eye-lines and his speed of running and things like that. So when it came to the acting itself, that was directed by our director Chris Sanders and the animation director, based at MPC in Montreal.
So the difficulty is that I wasn’t actually cutting a performance. We had previs and postvis, but that is not even timing-wise temporarily what is going to be, so that was a very difficult part.
So as editors we can actually try to help fill in those moments and say, “We’ve edited the scene and we want these things to happen.” Or “We want him to be emotionally set here at this moment.” So it’s quite different. I’d rather like cutting performances.
HEINZ: I worked on The Jungle Book for a while and in that film at least they had the audio recordings from the actors. So with Apes we had the motion-capture performances to cut with; with Jungle Book we at least had the audio to sort of make a radio play first. With this film, we had none of that. We did have Terry on set that we could cut in but it was really just a place holder for the time being so we didn’t really get the dailies back for Buck until very, very late in the process.
HULLFISH: Did you guys have to edit a lot of that like an animated film? How much of a hybrid were you guys working with between animated and live-action?
HOY: I have an idea of what animation films are like … now – I heard from friends who are animation editors – and it certainly felt that way because each time a new version came in then we would have to make changes to the timing of the animation that came in based on the emotion that it had.
Sometimes we have to rework scenes around them, so we were constantly changing the picture, which meant we were changing sound effects and the intent of a scene, so it was just an ongoing process.
The picture continued to evolve through the entire post-production.
HULLFISH: You mentioned liking to edit performance — that drives a lot of when you make a cut – when a look goes from one person to another, for example. How were you guys determining that kind of stuff?
In the final movie, with the dog’s eyes there would be a little glint or a moment or twitch and you guys couldn’t have been seeing that until very late in the process.
HEINZ: It was very much an ongoing process meaning that when we cut to the dog, at first we’re using a shot of Terry, then maybe we’re using a storyboard and then sometime later we’re using a bit of previs or postvis but it’s not until we start getting the animation dailies back from MPC and the animators up in Montreal that we’re then having to go in and recut.
So we end up recutting these scenes with all these different iterations of the shots many, many different times. We would spend long stretches of our days — particularly towards the end of post — in VFX reviews and then after those VFX reviews we’d go right back to the cutting room and have to revise some scenes based around new shots that had just come in.
HOY: When you say “the glint in the dog’s eye” and when they turn and the timing of that — we’re within a certain amount of frames, but we don’t know about a glint or when the actual sled is going to flash by… the trail of snow is going to be CGI so we then have to set that particular timing at the very end.
We have an idea of that going in because in certain iterations we talk about it. So we expect to see that at some point. And so we actually have to sit through all these visual effects reviews and watch all the different iterations of these visual effects, so that takes a lot of time away from the cutting room too. So we have to do that and then go back to the cutting room. So it makes for a long day usually.
HULLFISH: Talk to me about the process of what dailies looks like on a film that has so much animation.
HOY: David had started on the picture quite a while before I did. He came on to edit the previs itself. I knew that David had this huge headstart, so when I came on to the picture, I thought, “Well, this is going to be the easiest job I’ve had. They’re gonna shoot — in physical production — these shots and I’ll just drop them in and I’ll go home at 5:00 (we all laugh). This sounds great to me.” Of course, that was nowhere close to being the truth.
HEINZ: I was fortunate enough to be a part of the process for nearly a year before we filmed anything. On a typical film you might do previs for a handful of sequences — maybe a large action sequence or something like that — on this film we did previs for almost every scene in the movie.
They wanted to very meticulously plan out where Buck would be in any given moment and then, some of the later scenes that are fully CGI, we did pretty extensive previs on all of that.
As we were cutting the film together if there was ever something that wasn’t captured with live-action photography, we were cutting in previs as place holders quite often and for some of the later sequences that are entirely CGI we had really no real dailies to work with except for previs and postvis.
HOY: When it came to actually editing the picture, we have dailies and obviously we have the human actors and we want to get the best out of our human actors and then where Buck fits in with that story, we slot those moments in.
If we had a background that they shot or a plate that they shot where Buck would be, we would use that or it would be a virtual background. But there are scenes in the picture that are completely virtual, like the fight where Buck takes on Spitz and Spitz goes away.
There was a previs of that but ultimately the animation director – with our director – decided there were certain changes they wanted in the fight once we saw what Buck’s character was going to be through the picture. There were a lot of things that changed within that scene to make it what it finally became.
HULLFISH: Having done some animation editing myself I know that there’s kind of a process where you’re looking at some pretty crappy animation. It’s not that the animation is bad, but it’s just not finished or lit or textured. Is it hard to edit with that because there’s so little emotional attachment to it?
HOY: I believe that as editors we can imagine what that might be. The biggest difficulty is selling it to people watching it who don’t know what they’re looking at. “Here’s your movie! Show it to the studio!” And they look at it and say, “Well, I don’t feel anything.”
I don’t know how many times we have to say this: “It’s postvis and it’s in a very rudimentary form.” So I think that’s one of the more difficult parts is — as an editor if I show it to an audience — if I have the REAL performances I can kind of gauge how people in the room are. At least I can get an idea of: “it’s not working there.” Or they reacted in a way that I could not have predicted, but with postvis characters that have no emotion, it’s very difficult to gauge the audience’s reaction, so am I second-guessing myself?
It becomes difficult because you don’t really have a defined gauge where you can say, “That’s really working!” I’m guessing it’s working based on past experience, but I have no way of knowing with any amount of certainty that it is. And that becomes difficult.
HULLFISH: David, talk to me a little bit about the pacing with emotions because if you’ve got a character like that dog that — in their final animation is so emotional — but during early post, you’re looking at postvis which isn’t like that — does it mean that when you’re cutting the postvis you want to cut quicker because it doesn’t appear that the shot has the emotional weight to sustain more time?
HEINZ: Yeah. I think there is a tendency to try to cut quicker, but Bill and I had conversations about that very thing. I think Bill — with his experience particularly on the Apes movies — was wise to caution about cutting too quickly too early.
Obviously, the animation is a ton of work and working on every single frame of animation is a ton of work for an entire team of people so we do try to limit the amount of work they have to do, but that being said, we don’t want to paint ourselves into a corner with any given shot where we haven’t alotted it enough time to breathe, and then the dailies come back and it feels too brief.
To touch on something. You guys were speaking about a minute ago in terms of all the rudimentary iterations of a shot leading up to the final product, we had to test-screen the movie a few times and we had to do that pretty early on with some pretty rough previs and postvis in there. And I was really struck by the fact that the audience kind of went with it.
At the beginning of any screening, we said, “Here’s what you’re going to see and here’s what a finished shot is going to look like.” So the audience was kind of ready for that, but it was very encouraging early on that it felt like audiences slipped right into the story and started feeling the emotion of Buck and his story right away.
HULLFISH: What would a bin for a scene look like?
HOY: For a normal picture you would have your bin of dailies from physical production. We don’t really have a bin for the animation that comes in. We have a visual effects editor, and they keep track of all that.
But we do have — on our timeline — the different iterations from the moment that it’s as simple as a plate and then we take the action of Terry Notary who gives us the eye-line. So then we rudimentarily comp that in there. That becomes the basis of what it is.
Then the next iteration will probably be replaced by postvis where they put a dog in there that’s almost two-dimensional. Then we continue on from there so our timeline would be 13 15 stacks of video tracks.
There are many, many more iterations than just those versions that exist on the timeline but those are the ones that are pertinent to the actual shot itself as far as timing, as far as acting is concerned. So that becomes more like our bin.
HEINZ: It can get complicated with so many video layers, but having done it with all these big visual effects films before and working as a visual effects editor we were able to keep it all straight.
One thing to keep in mind is that when you get a bit of animation back from MPC or even a bit of postvis back, we’re getting them shot by shot. It’s not as if they’re delivering dailies for an entire scene: “Here’s a whole scene of animation dailies.”
It would be amazing if they were able to do that but it’s just not the way it works. They have to give us shots as the animators finish them, so after we finish up a VFX review, then Bill and I are going back down to the cutting room and recutting stuff. We’re re-cutting one new shot surrounded by three other shots that we’re still waiting on new versions of.
It becomes a pretty complex process where we’re very reliant on the visual effects editors on a film like this.
HULLFISH: When I was thinking of what the bin looked like, I was thinking more when you tried to start cutting a scene from scratch, how similar was the process to a live-action film? Or did you always base it on the previs edit?
I’m thinking they shot something and then you have a bunch of previs that’s also in the bin for the stuff that doesn’t exist in the shot.
HOY: In the cases where we had some previs, we would slot that previs in there and in a lot of cases we just had an outright shot missing so we would just create a banner that said: “Buck runs off-screen left” or “Buck looks to his left and then trots off left to right.”
Based on just these surrounding physical actions that we have already on film then you can kind of look at it and think, “OK, this is probably what’s going to happen.”
Everyone has their own concept of a description. When I say, “It’s a closeup of Buck and he gets excited and runs after a rabbit.” How big is that closeup? And when he runs off to the left, is that three-quarter left or straight left?
So ideally we could have some kind of image that we can manipulate in a very simplistic way. So if I can make some kind of template it saves a lot of words and a lot of emails back and forth.
Getting back to what David was saying: on the Ape movies WETA went about it quite methodically in the sense that they devoted a crew to certain scenes. They would concentrate on a certain scene and when it was almost done then they would go and concentrate on other scenes. That’s how it went for the most part, which meant that the dailies would come back on that scene and we could see all of the shots for a scene at once. That helped because we could tell when things weren’t working.
Whereas with this picture we get one shot maybe two shots that aren’t even together so we can’t even make that edit itself. So that in itself makes it very difficult. Part of a scene is postvis, part of it is dailies, part of it is unfinished animation that we get from MPC. It’s all temporary.
And by the way, Steve, we’re not only carrying 13-15 video tracks, we’re carrying 21 sound tracks – 5.1 tracks; stereo tracks; dialogue tracks. So when a few shots come in we have to adjust everything.
HEINZ: Adjust everything all the time.
HULLFISH: David, after spending a full year on previs, how much of a template was that compared to your final movie?
HEINZ: Some sequences more than others. Like any film it evolves and changes as the film is shot and reshaped and rewritten. Kind of the heart of it remained. I think the Spitz fight in the middle of the movie started in previs and became a pretty good framework where we ended up there. The avalanche was pretty extensively previs, and I think ended up in the ballpark.
But for a lot of the emotionally-driven scenes, the previs was kind of just an approximation. There’s no way to know how you’re going to block out or cut a scene like that until you get the actors there on set and you work through it on the day. So a lot of that stuff was sort of just place-holders.
But what it did afford us was an opportunity to almost screen a full cut of the movie before we shot it. And what was nice about that was: as more people were being hired on the film — or they were casting the film — they were able to show them parts of this previs and now all of a sudden everyone’s speaking the same language. They understand the intent. It might be an approximation of that intent but everybody now has a basis for what we’re doing in any given scene or sequence, so I think it was really helpful in that regard.
HULLFISH: Structurally, because you have the ability to kind of see the whole movie play out — which is very rare unless you’re doing an animated film — were story decisions made during previs that wouldn’t have been made in a film that was shot traditionally?
HEINZ: I would say yes. I think so.
The sledding montage, for example, where Buck is learning to become a sled dog. Because we had conceived that as a montage in previs when we went to shoot it they just picked up a lot of little pieces. They weren’t shot as full scenes that we then trimmed down and honed into this montage. It really set out to be a montage.
So Perrault lifting Buck over the cliff — for example — there are only two setups there. They knew it was just going to be a quick shot or two and then we were into the next thing. So in that regard, I think it was helpful as well to know: “OK. We need a sequence where Buck becomes a sled dog but we need to do that rather efficiently. Let’s look at doing this as a montage.”.
HULLFISH: Near the end of the film, after Harrison Ford’s character goes off into the wilderness, there’s a bad guy that starts to pursue him, but there’s quite a bit of time with Buck and the Ford character off in the wilderness before the bad guy comes back. Was there any thought – or had there been other scenes – where you track the bad guy getting closer? Or was the discussion: “We want to just be with the dog and Harrison, and we’re not going to worry about the bad guy until he shows up?”
HOY: I think the only scene that was missing was the bad guy finds the canoe that’s been wrecked and we know he’s on the trail. But I think in that area where Buck and Thornton go on this journey together the feeling was that we wanted to concentrate on this adventure of a lifetime.
So finding the canoe broke that rhythm up of them enjoying the wilderness together. And then there was another introduction there to the White Wolf. So what do you give your time to? And the feeling was that we didn’t need the bad guy to track him because he wasn’t really giving us any more.
Ultimately we did feel that Buck and Thornton had this adventure of a lifetime and that Buck did find his wild self. To concentrate on that was the right thing to do.
HEINZ: When we went to preview with an audience, we purposefully left that canoe scene out knowing that if the audience feedback came back as: “Where did this guy come from? He just came out of nowhere!” we knew we had that scene to put back in, but we thought we would test it without and see if the audience went with it. And it seemed like they did.
HULLFISH: That’s a very interesting idea that I’d heard from other people too — that when you test something you want to test it OUT instead of testing it IN. And I’ve even heard that with editors just screening it for a director. “Hey, I’m going to try cutting this out and seeing if the director even notices.” Directors are pretty invested, so they’re probably going to notice, but you can cut things out and see how they’re gonna play instead of maybe ASKING to cut it out.
Has either of you tried either one of those tactics?
HOY: Actually I took something out on this movie. I dropped something because I felt like it should be out. He watched it and liked it. I asked, “Did you miss anything?” He said, “no.” He was shocked that it was pulled out and he didn’t miss it.
Still to this day I’m surprised that certain parts of any picture — when you take things out — even scenes that I really love the acting and the setting. You think that you’re getting so much out of the scene but you take it out and things come together and the bigger aspect of the movie works so much better.
You’re sacrificing this beautiful scene and you don’t want to lose it. But it is one of those things where to this day I’m surprised about what can be cut from a movie.
HULLFISH: Have either one of you guys done one of those slash-and-burn edits where you cut to the bone. You take it too far but now at least you know what’s critical or maybe some interesting scenes are juxtaposed accidentally because of cutting another scene.
HOY: We have done that on different pictures and yes, it was like that. You find it really works well in certain places and in some places: “No. I really miss that.” So then we actually have to put it back in.
With some directors, we start fat and then we trim little by little, then at some point we say, “Let’s just cut this right down to the bone and see what happens. Just to watch internally. Nobody should see this but us.”
We’ll do that and you discover the moments you miss and really want back in. Sometimes something you drop works really well, except you lost some important information so then you try to see how you can put that important information — or a glimpse of what that idea was — somewhere else.
It could be an ADR line. It could be just a shot that was in that scene. We don’t know until we experiment. Then you can see what you actually can lose to make the whole a bit better.
HEINZ: One example, Steve, from Call of the Wild where we did that is after Buck met the White Wolf, we had a really beautiful sequence where Buck was taken to the Wolf den for the first time and was kind of introduced to the entire pack of wolves and at a certain point tried to just lift that out.
I thought there would be no way we could live without that sequence. I wasn’t ready to lose it, but sure enough, that was one of those lifts where as soon as we pulled it out everything else just came together around it.
But to Bill’s point, we knew there were a couple of little moments from there that we really wanted. We wanted to establish the Wolves’ den and the tree where they lived, for example, so we took two or three shots from the start of that scene and put them into the end of a montage that comes later in the movie. So we retained that beautiful establishing shot, but sure enough, when we lifted that scene right out everything around it just seemed to go together beautifully.
It’s kind of like in basketball — you see a guy take a crazy shot and think, “NO NO NO!” and then it goes in, and you’re like “All right!”
HULLFISH: I want to stay on the discussion of story with a movie that you’ve prevised so much. Did less structural change happen in post than with a typical movie?
HEINZ: I don’t know that less of that happened to be completely honest. But I think there were some story problems that we solved early on which proved to be really helpful later.
Like anything you do as much meticulous planning as you can, and then once it’s filmed and you get footage back the film starts becoming what it’s going to be and evolves over time. Especially a film like this where there are large sections of the third act that are entirely CG, we could sort of write and rewrite those right up until we had to finish them with the animators. So it was absolutely an evolution.
I think the overall structure of the movie is set largely by the book anyway, and with the source material being such a classic there would be no great reason to deviate much from that anyway.
HULLFISH: Did both of you guys read the book or did you intentionally stay away from reading the book?
HOY: I did read most of the book. (Dave and Bill laugh) I’ve worked on a few movies that were adapted from books. When people read books — and I know for myself that I have ideas that I love about the book and sometimes might get tied to.
I remember the first picture I worked on was based on a book and I went out and read the book and I studied it, and then when I read the script it was not even CLOSE to the book. So you have to divorce yourself from the book. It wasn’t out of laziness that I didn’t read the book. I wanted to revisit it just to get the feel of it, but the book is pretty brutal and we weren’t making that movie. But the actual spirit of the book, I think we’re adhering to.
HEINZ: At least when I went to school, it was required reading so I read it in school, but I hadn’t picked it up since I was a kid. I reread it before I started on the project and I was reminded of just how vicious and brutal the book is.
It’s amazing and it’s a classic but rereading it I was reminded of just how sort of intense it was and I think the intention all along was to try to make a film that was palatable for families and I think they were aiming for a PG rating all along so we knew right away that we were needing to adapt the spirit of the book more so than the moment to moment intensity of it
HULLFISH: So there are definitely some challenges in working on a film with this much CGI. Are there advantages?
HOY: As difficult as it is on us personally to work on fully animated movies like this — or fully CG movies when you have a CGI animal — I feel like I’m more a part of the storytelling process. When we have the edit in front of us we can actually talk with the director, the sound people, and so it feels like we are even more collaborative than we are with a standard movie.
It feels that way. I feel pretty collaborative on just straight storytelling movies but this way it actually feels like you can be on the set and say, “Let’s move the camera over here.” Or “Let’s go closer here” or “Let’s have the camera moving you in the shot.”
We can suggest those things and we can show it on our Avids and just give them a blueprint to go on — and if nothing else stir ideas. Like, “I have an idea for this. We’re having a problem with this section of the movie. Let’s try this.” Then I can lay something out so they can see it, and then that germ of an idea becomes a bigger thing.
It feels very collaborative that way. It feels like I’m really deeply involved in the movie. That’s the part of it that I do enjoy very much.
HEINZ: Being a part of the process so early is such a rarity and a really great opportunity an editor to weigh in early on the storytelling.
I ended up shooting virtual camera on a lot of scenes that became previs scenes that then became the basis for how we were covering and filming sequences in the finished film. So in that regard, it was a very, very rewarding process. You have ideas. You try them. You run them for an audience. They work. There’s something really, really gratifying about that. Being involved so deeply in the storytelling is a dream come true. It’s what I love to do. So it was a joy.
HULLFISH: You mentioned a virtual camera. Can you describe the technology of what you were doing with the virtual camera? Was it actually like running a camera inside of a virtual space?
HEINZ: Essentially, yeah.
The way we created a lot of the previs for scenes that had humans in them is we brought in human actors and we put them in motion capture suits. Chris Sanders, our director was then able to block out scenes with the actors. That motion was captured and put into a previs environment.
So now we had the movement of the actors and the blocking that Chris had set up for any given scene, but we didn’t have any camera angles. So within that environment, we then needed to take a virtual camera — which is not much bigger than an iPad. It’s a handheld device where you go in and you can set camera angles on any given piece of animation.
So it was from those angles that I would get dailies back in previs form and cut those together. That was great because I would cut something together I’d say, I think we need another shot here — a little bit more coverage — whatever the case may be. And we were able to go back with that virtual camera and just shoot what we needed.
So that was very gratifying. The motion capture stage is right next to my cutting room in the previs phase of the movie.
HULLFISH: When William was talking about having input I was thinking that on a typical shot film the editor can’t control the pacing of delivery or performance within a shot (maybe with a split-screen or a timewarp effect) but you guys could actually say, “I want the pacing inside of the shot to change. I want the reaction to be quicker from the realization to the dropping of the eyes or the looking up” or whatever.
Stuff could actually change interior to the shot.
HOY: Certainly we take physical shots and put them in places where they weren’t intended and that’s how we would make it work. There are little tricks to manipulate the frame even if it’s shot, but the freedom of doing it with CG characters — even on “Apes” we could take the head if he’s looking the other way and have the eyeline be correct and so it was like three-dimensional chess.
HULLFISH: David, when you were working on the previs, how much audio stuff did you have? Did you have a lot of dogs whining and barking?
HEINZ: We had a full soundtrack. It was pretty rough and very temp, but we had sound effects and music in for the entire film.
We had a music editor — Ted Kaplan — who came on and helped us with the previs, so the previs was very much a viewable early cut of the movie.
HULLFISH: What do you guys think about doing that on a regular non-CG movie? What would you think of that concept of trying to previs a straight shot movie that heavily? Good or bad or a waste of time?
HOY: I think it’s a waste of time because in a virtual world you can do all these different things and once you get on a physical set you actually have to shoot it, It becomes a different thing.
We have had scenes in a lot of movies that were previs just so the production designers, stunt people, everybody could get on the same page about: “The camera’s going to be here. And this is the type of shot I want.” But when we get the actual dailies back from a scene that had been completely prevised, it doesn’t look anything like that.
I’m working now with a director that doesn’t like previs but he knows the value of it because it’s like a more detailed storyboard, if you will.
HULLFISH: Just a thought experiment. I saw a video on This Guy Edits with Dr. Karen Pearlman, who teaches film, and she makes all of her students pre-shoot their movies and edit them together to see what the story problems are gonna be. You learn things when you’re shooting cheaply and you’d try not to repeat those mistakes when you were working with a full crew.
HOY: I think if you want to do that, you should take your iPhone and get your friends and shoot it and block it out. It’d be a lot quicker and you wouldn’t have to go to all that time and expense and you would actually have movement and you could actually see if it’s going to work or not. At least the blocking would be on the same plane as a physical set.
HULLFISH: So you think there might be a value in doing that if you shot it instead of prevised in a computer?
HOY: I do think so. From what I know on some of the movies that I’ve worked on — the actor got his actor friends together and they walked through almost the entire movie so that he understood what and how he was going to shoot it.
He actually had it blocked in his head. “Like, When I confront this person I want to be here.” So he has it now blocked out. And even though you know mobile phones with cameras didn’t exist at that time he had a really good idea and it just made him more prepared when he got on the set: is this gonna work or not?
It is important to know going in how you’re going to shoot something, however, you find a way to do it. If it becomes so inexpensive that you can do a whole previs and do it virtually, go ahead. Whatever tool helps you.
HEINZ: I think as great as previs is as a tool and as much as it’s evolved in terms of its look, it still has a limitation in terms of performance. So performance is always going to guide the story — drive the story — and really only actors can bring that to the equation.
So previs is a great start, but I think Call of the Wild is the exception to the rule. I don’t think prevising an entire film is largely helpful to many films.
Thanks to my guests, William Hoy, ACE, and David Heinz.
The first 50 interviews in the series provided the material for the book, “Art of the Cut: Conversations with Film and TV Editors.” This is a unique book that breaks down interviews with many of the world’s best editors and organizes it into a virtual roundtable discussion centering on the topics editors care about. It is a powerful tool for experienced and aspiring editors alike. Cinemontage and CinemaEditor magazine both gave it rave reviews. No other book provides the breadth of opinion and experience. Combined, the editors featured in the book have edited for over 1,000 years on many of the most iconic, critically acclaimed and biggest box office hits in the history of cinema.