The long story of Doktor Fang...
This is the long and overly detailed story of how 'Fang came to be. If you've ever worked on a movie then there's nothing here you don't already know. But if you're a student or would be film maker there might be something here that's worth the read -- something that might save you a few hours, a few bucks, or a few headaches. This is all more or less as I remember it, but I've taken some dramatic license with the quotations, so no one I mention should be held accountable.
Doktor Fang began as a conversation with George Calamatas at Romelo's Cafe in 1999. George was looking for something to animate on his Amiga computer and showed me a picture of some early toon-shading tests done with Lightwave. "German Expressionism" popped into my head. At that point I was in a post-production holding pattern on a short film called "Tales from Space" which was essentially an homage to the aesthetic of sci-fi films of the thirties, forties, and fifties. German expressionism wasn't being heavily explored (apart from Guy Madden) and stylistically it would mesh easily with toon shading. George was looking for something between two and five minutes. Ten minutes of talking later I'd dragged out enough of the Kafka clichés to have a workable premise for a story. The name "Fang" was originally spelled with two dots over the A, to make it seem more German.
I wrote most of it in the next day or so and then pounded my head on the wall for six weeks trying to figure out my third act. When the light-bulb finally came on I finished the script in about an hour, polished it the next morning and George and I went over it - it was a bit long (about sixteen pages), but there wasn't really anything we could cut. I was briefly employed for the pilot of "Live Through This" in the transport department, and that's where I did my storyboards. Meanwhile George made some preliminary designs -- excellent designs, by the way.
But the project died: there was no money, and the energy required to animate a sixteen minute project of this scale on a 1998 Commodore Amiga would have kept the machine rendering for years.
Years and years passed. Another friend, Franco Zoccali, asked an animator friend of his to do a rough budget. Including development, it came in at about $800,000. Even allowing for contingency padding (and I don't think there was any) it was completely impossible.
Early in 2010 Telefilm rejected one of my feature-film projects for the second time, and the producer Melissa Malkin said "you have to make an award winning short film to show them you know what you're doing." I had nothing that was ready to go, or which hadn't already been rejected, so I dusted off 'Fang.
It took a month of seventy-hour weeks to write the grant proposal. The project would no longer be animated, but would instead be live-action green-screen with 3D backgrounds. This meant creating sample images, building steam-trains, and getting up to speed on production techniques; lots of time rendering test images; days and days and days...
There was also the conceptual pitch - what makes this project worthy of a grant? This is the part I've always hated, because you have to describe and illuminate things which (you would think) should be obvious to anyone in the arts business (Art is a mirror revealing the audience to itself -- you tell me what it's about). In fact this part of the process turned out to be a blessing in disguise. When someone asks what a film is about the answer is invariably less than the work. But any writer, director, actor -- any of the creative credits in a production - should be asking themselves this. It's a good place to start rewriting, to start your actors (who will inevitably have a different idea than you, anyway), and to start your creative team (DOP, Production Designer, Wardrobe Designer, Musician, Editor, etc.), and being able to articulate that concisely is important. It wasn't until I wrote the proposal that I realised "Fang" was about "the purpose of dreams"; that however you might interpret your dreams, their purpose might be something else; that a dream's meaning and its purpose might be two different things.
At least that's what I told the Canada Council.
The grant requested was for established Artists ($60,000). Melissa was sceptical: "I don't think you'll qualify as established". "Why not? I've sold scripts," I argued back. The budget was $60,000.
I put the package together and shipped it express out the morning of March 1st.
I was booked to work in transport on a feature starting sometime in April. Jacques Bernier and I went to see "Avatar", one of the worst films I have ever seen, and which had already shattered box office records (I guess Art is a mirror). By the time the movie was over I had my next project: a 3D Screenplay!
In fact the 3D Screenplay had dated back to 1997. William Gibson had once described his experience writing a screenplay: it's a work of art, until someone makes a movie out of it and then it's just an artifact and the movie is the work of art. How could someone retain the "artiness"; making the screenplay itself into an object (as opposed to a plan for an object). In 1997 the technology was too complex to execute on a home computer (the story of my life) - people were still transferring data on floppy-discs. But by 2010 anything was possible! A quick Google search and I had everything I needed.
It took about two weeks to convert two short scripts ("Doktor Fang" and "Pudovkin") to anaglyphic 3D, mainly because I was using a computer that was five years old and choked with Windows system artifacts, and because I restarted the project when I figured out what I was doing. I ordered some 3D glasses and set up a web page. The 3D Screenplay was a joke of course, but it was a finished joke - a completed work. Something that stands alone. I realised it had been years since I had had the pleasure of finishing something that didn't need to move onto another stage.
The day before I was to start my transport gig I got a call from Ian Reid at the Canada Council: "You won't qualify for the $60,000 category." "But I've been doing this for quite a while now -- people pay me to do this," I argued. "You won't be eligible," he said "But you would be eligible in the emerging artist category." Emerging Artists have a ceiling of $20,000. "The only way I could do that was if I didn't pay anyone," I pointed out. "You won't be eligible for the Established Artist category," he repeated. "What should I do?" I asked, throwing myself on his mercy, "The budget is for sixty-thousand and I don't have time to revise it down to twenty-thousand." "Change your budget summary page," he suggested "to reflect the new budget and then you will be eligible for the Emerging Artists category." Of course this wouldn't guarantee me a grant, it would just make me eligible for a different category -- a category forty-thousand dollars less than my budget. Even if I got the grant (highly unlikely, given that the budget was now clearly inadequate) how would I make the movie? It took twenty minutes to make the changes and email the revised summary.
I forgot about the grant immediately.
The movie business is basically about standing around waiting for someone else to make a decision about your work. When a novelist completes a novel, they have a novel (even if it's never read, it remains a stand-alone work), painters have paintings, musicians have music (assuming they can play an instrument or use a computer), but screenwriters and architects have drawers full of uncompleted plans -- ideas waiting to be financed, altered, revised, and frequently "autered" by someone else. Making the 3D Screenplays had given me a taste for finishing things. The week I started my transport gig I got an email advertising comic-book software. I've been reading comics seriously since the eighties (Love and Rockets is still the best book out there), but the form was always a mystery to me. In 2007 Drawn and Quarterly opened a store around the corner from my apartment and I've been a regular customer ever since. I thought, they must have something on making comics! That Saturday morning I wandered in and found Scott McCloud's "Making Comics".
"Making Comics" is one of my top ten most useful books. Even if you have no interest in making comics the book is and illuminating insight into narrative structure and semiotics. In fact, anyone interested in the arts should read this book: it's a page turner, genuinely inspiring, and if you do want to make comics, the most informative book on the market. I had a number of projects I thought might be comic-book worthy, but they all seemed to be a little ambitious for someone who had never made a comic before and who couldn't draw. I was rereading all my Love and Rockets books when I came across a 24 panel two-page masterpiece by Jaime Hernandez called "Son of Butt Sisters". I realised the smart thing to do would be a one page story, to get a feel for the medium, figure out some problems, and find out if I had the stamina or temperament for the genre. And -- most important of all -- I would have a finished piece. No matter how atrocious the work was, it would be a complete work of art. But how to do it? Oh yeah, that comic book software.
In movie transport departments the work week is seventy to eighty hours. You start first (picking up actors and technicians) and finish last (dropping off actors and technicians), which means doing things on a week-night is a bit difficult. I found a few hours during the week to try out some techniques using Manga Studio 4 EX, and the results were encouraging. That weekend I drew my first page ever. Yes it has shortcomings, but it is also finished -- more or less. In fact over the following weeks I'd try out stuff, draw more pages, and learn something about the medium. Everything I did had significant problems -- but they were finished!
I wrapped the show I was on, and was in standby mode to start another when I remembered the Canada Council application. It had been over three months and I hadn't heard anything. Those assholes hadn't even had the courtesy to send me a rejection notice! I shook my fist at the sky: "I'll show you, you fuckers!" I prepared to write one of my famous "letters of complaint", but first I needed groceries. At the front door I found an envelope from (you guessed it) the Canada Council. Enraged, I ripped it open.
"We are pleased to inform you that you are the recipient of a grant for $20,000."
"Shit! Now what am I going to do?!"
Producing (in my limited experience) is a like fighting a war on multiple fronts: you send your available resources in different directions trying to gain ground while heading off potential problems. It was September and I wanted to shoot at the end of October, which meant solving problems quickly:
1 - the Budget had to be revised in order for the Canada Council to approve and release the funding. This meant figuring a way to cut out $40,000.
2 - Personnel: cast and crew had to be found - people who knew what they were doing but who would still would work for a nominal amount of money. I couldn't afford union rates, but since I was in studio I could offer good working conditions, food and fixed hours. The budget was so limited I'd never be able to get more than three days of shooting, and since I couldn't pay overtime I had to restrict the hours and stick to those restrictions. No one was going to quit their job to work on this project, but if they didn't have anything they might come on board.
3 - Studio and Equipment: no point in hiring people if there's no place to work. We needed a green screen cyclorama that was pre-lit, in a studio that came with on-site equipment rental.
4 - Art Department, Props, and Wardrobe: all that stuff that you see actors touching, wearing, or sitting on.
5 - Post Production: how does the show get finished? Someone has to edit, score, and replace that green-screen with backgrounds. This was obviously going to be a time consuming process, so I had to find someone who was available 24/7. Oops - that would seem to be me.
6 - Administration and legal: insurance, workers compensation, taxes, accounting, etc. Always the last thing anyone wants to deal with, and always the first thing to screw you up. I'm terrible at this.
Two people came on board immediately: actor Matt Holland, who played most of the supporting parts and suggested some other actors; and Jacques Bernier, who photographed it. Jacques has been a friend of mine ever since he photographed "Tales from Space" in 1997. It became obvious that the only way to get the project finished was if we shot 40+ set-ups a day -- one camera set-up every fifteen minutes; in other words, fifteen minutes to block, light, rehearse, and shoot each shot before moving on to the next angle. Jacques went looking for a studio and his crew.
Since I was now in charge of post-production I needed to figure out what the post-prod pipeline would be. In other words, once the movie was shot, where would the footage go and how would it be treated? After consulting with some very knowledgeable people I settled on the Adobe CS5 Production package. This package includes After Effects, Premiere Pro, and Encore (among other things) and uses Dynamic Link to keep everything together. I bought the package, some instructional literature, and invested a lot of hours doing tutorials so I'd have some idea of what I could get out of the material.
The post-prod choices dictated the camera system we would use. Jacques is a fan of the RED, but I needed something I could off-load onto hard-drives without the need for a data wrangler; a format that an idiot like me could wrangle at home; a format that wouldn't overload my desktop system. Experience has taught me there's almost always an "expert" around who has an "easy" answer to everything -- gadget guys who like the toys -- and almost inevitably when the time comes to implement the "easy" solution they remember a key piece of information that either blows out the budget or your schedule or your patience. Most of the time this is someone who's agenda is only peripherally related to the completion of your film; the classic example is the Director of Photography looking for something they can put on their reel, renting lots of lights, and spending hours making adjustments while the shooting schedule falls apart. Thankfully Jacques is most definitely NOT that guy. After weighing the pluses and minuses we went with the Panasonic HPX170 -- not Jacque's first choice, but the most compatible with my budget and pipe-line requirements.
Jacques was also touring studios, looking for something with the facilities we needed at the price we could afford. Lighting a green-screen takes time and energy and a big ladder. In 1997 we'd shot "Tales from Space" at Michel Trudel's Studio D. It was a great space with a high ceiling, but all the gear was now off-site, and the studio was already rented, so going back there would require more time and resources than I could afford (a shame -- for small productions, it really is a nice space). Jacques found another Studio D at Cinepool. The green-screen was only twelve feet high, but it had a 35x25 foot cyclorama and was pre-lit; adjacent to the studio were make-up rooms and a common area and the deal included all our lighting and grip equipment, and a camera package from Department Camera. By the time I booked the studio my shoot date had moved back to mid-November. The studio was already booked for my ideal dates, so I rolled back to a Friday-Sat-Sun arrangement.
Now that I knew the studio dimensions I was able to model the space in 3D and do storyboards that were camera accurate - i.e. heights, distances, focal lengths all exactly what they would be on set. The space limitations of the studio made this step absolutely critical: the limited ceiling nad floor space determined all the angles. There was, litterally, no room to manouver! I re-boarded the entire show and sent the boards to Jacques. We were still at 140+ setups, and some of these required doubling up -- many shots featured an actor playing twins.
The studio lock meant other things had to be dealt with immediately. We still didn't have a lead actor. Matt had some recommendations, but the first two fell through, and the third wasn't available when the studio was. Previous commitments meant rolling our shooting date back or forwards was now impossible. Fortunately I did have The Woman cast: Gillian Ferrabee. Matt had recommended her and I'd seen her on several sets over the years, but the only thing I'd seen her in was short film on the internet and she wasn't very good. I met her, she liked the material, was available, and was an all round nice person. I broached the subject of this film I'd seen her in ands she got all embarrassed. She gave me some titles of DVDs and I scooted off to view them. I hired her before the movies were out of the DVD player.
Jacques had found an Electric and Grip team, but that still left sound, unit, costume, make-up, art department, and so on. Meanwhile I was dealing with bureaucracy, casting and everything else, and was falling behind. I'd been bumping into Esteban Sanchez for the last two years, in his varying capacities as a production coordinator. When I told him about my grant he said "Dude, if there's anything I can do..." I called him.
Esteban jumped in immediately. He covered all the bases I hadn't. He found a coordinator (Eva Quintero), Assistant Director (Renee Carre), Caterer, Vania Khandjian (Costume and Art Department), and Sebastien Moise (unit manager). Last, but by no means least, were Tobias Haynes (sound recordist) and Jose Bernard (boom). Tobias brought along sound equipment generously loaned by Don Cohen. In other words, Esteban saved my ass.
We still didn't have a male lead, so Esteban set up a meeting with Andrea Kenyon at Keyon Casting. Andrea offered to do full auditions, but I decided to just look at audition tapes already in her library. It's faster, less intrusive, and, unless you're auditioning for Hamlet, one audition tape is more or less like another and you can be more impartial if they're not trying to perform your material. We saw some excellent possibilities, and one definitely not (if you're going to audition for anything, learn the dialog -- don't stand there with the script in your hand), but no one who "popped". One of the last tapes I saw was a guy with a beard who looked homeless. I was looking for someone who might pass as a clean shaven young scientist from a 1900 Vienna, and Brent Skagford looked like University dropout from 1990. But something about his mannerism said "Fang". "Will he shave for the role?" I asked Andrea.
I met Brent the next day at a cafe near his house (Brent, Matt and Gillian all live within a half-mile of my apartment). He was clean shaven, with a crisp, short hair-cut and half-rimmed glasses. He looked like Aticus Finch's kid-brother, a scientist from the Manhattan Project, or one of those guys that always gets killed by mistake in an act of revenge. As it turned out he hadn't shaved for the role, he always looked this way. Perfect.
I met 1st Assistant Director Rene Carre at Romelo's cafe about a week before shooting. He looked vaguely familiar, and we realized simultaneously that we'd both worked on "Black Robe" (1990); Renee had run the Java WWIV bulletin board where I played Trade Wars for several years. Renee already had some breakdowns and a schedule in place. We moved some scenes around, and everything looked very good. Then he said, "I think you're looking at fifteen to sixteen hours on your first day." "We wrap after twelve hours no matter what," I told him. I could see he was sceptical -- and rightly so. If I had a dollar for every time those words had been uttered in the movie business I wouldn't have to worry about overtime. But it was a promise, and not only to the crew but to myself. And it was an easy promise to make -- we were a week from shoot and I was already sick of my own project.
Hair and Makeup was done by Olivier Xavier, who also has a long history in prosthetic makeup effects. He was building my masks, and we spent an hour trying out designs. I say "we", but he had most of the designs down and I just had to approve them.
My first good day occurred at our costume fitting at Noccoletta Massone's costume shop with Matt and stunt coordinator Jason Cavalier. A week or two earlier I had met with ACTRA rep Shannon Joutel to figure out how I could pay the actors as little as possible and still be use them legitimately. Shannon had read the script and had a number of concerns -- not about money (in fact ACTRA is very flexible about that at the low budget end), but about the conditions the actors would be working under: "They're in a forest at night and it's November - do you have adequate facilities to keep them warm? And I don't mean a car for them to sit in between takes." Most of her concerns went away when I explained that it was all studio and they had a greenroom at their disposal. But the subject of "fight" came up: "Will you have a stunt coordinator?" I had a certain type of fight in my head and I didn't think it was necessary, but Shannon's question raised enough of a doubt that I contacted stunt coordinator Russell Yueng and threatened to tell people about his early days in Montreal cinema if he didn't agree to do my movie. Russell was clearly frightened (well, politely pretended to be), but he was already booked on another gig and recommended Jason. I called Jason, described the situation, and he said he'd see what he could come up with. We were supposed to do a rehearsal after the wardrobe fitting but somehow the wires had gotten crossed. Jason had to get back to the country to gravel his driveway, and Matt had something to do, too. "Why don't we just do it here," Jason suggested, as we wandered past a park in the Montreal east end. He showed Matt a few moves, and in minutes I was laughing my ass off. "This thing might turn out to be a piece of shit," I said, "but at least we'll have a good fight scene."
It was also at the costume fitting where I met Vania Khandjian. "She normally works as an assistant," Esteban told me, "but she wants to try the Art Department or Wardrobe." My first reaction was sceptical caution: assistants are notoriously enthusiastic and regularly incompetent (that's why they get fired so often). Vania was smart, enthusiastic, and had good ideas -- better ideas than me. We tried lots of stuff, looking for contrast between characters (they can't dress the same), character consistency (would he/she wear this?), nothing green (green on a green-screen = bad), and must work in monochrome (textures and tones, not colours). The fittings went over the estimated time. I had an idea of what I wanted, but finding it took twice as long.
Two days before shooting Vania and I toured the local prop suppliers. We started in the east end and worked our way across town. Vania had a list of props and we worked from 8AM to 4PM filling it: hammers, watches, giant wrenches, drafting table, etc. We finished the day at Cineffects where Ryal Cosgrove had generously donated a treadmill, turntable, and green-screen paint. While we were there we saw some welding gloves and leather aprons, and Ryal lent us those, too.
I built the machine helmet myself. Vania and I had seen some interesting ideas in our props search - most of them left over from kids sci-fi tv shows - but nothing prêt-a-porte. The design was brutally simple: enamel bucket painted black, covered in door-stops, wing-nut bolts and wire. The "ideas" were developed in an afternoon of despair rolling a shopping trolley up and down the isle at Home Depot. All the buckets had colour enamel on them and I wanted something metallic. I asked someone with an apron if they had anything else and he didn't think so. I put the bucket on my head -- trying it on for size -- and then left. Very Ed Wood. Good therapy, too. I also built the wooden support for the switch Fang throws using pieces for a porch railing. It's interesting how these situations get solved: what I had imagined as a sophisticated piece of scientific equipment was reduced (by necessity) to a painted bucket and some bolts. This probably wouldn't have worked on the set of 2001: A Space Odyssey, but it was unexpectedly appropriate for Fang. Make your limitations your strengths! The props we really needed we found, what we couldn't find we created, and what we created redefined what we'd found.
Rehearsing with Gillian and Brent was a pleasure. I'd booked four hours in the local YMCA conference room. We did a read through, talked about the characters, rehearsed the key scenes, and that was it.
Judith Weston has written a great book "Directing Actors" which I wish I'd had access to twenty years ago. Rehearsal isn't about learning the lines -- actors are expected to know their lines the way a taxi driver is expected to have a driver's licence. Rehearsal is about discovering and agreeing on the character arc. Writers (and, unfortunately, many directors) use adjectives (slowly, silently, angrily) and actors need active verbs (study, sneak, hurt). The verbs describe motivation: why am I moving slowly? Because you're studying. Why silently? Because you're sneaking. Why angrily? You want to hurt -- the difference in meaning is night and day.
Mapping a scene into a progression of active verbs for each character is a good way of demonstrating motivation, and forces you to examine the nuances of that scene: is a given line a continuation of a motive (use the same verb), a new idea (find a new verb), or an anachronism (think about re-writing or cutting it)? You can discuss verbs with actors, because they are the articulation of motivation: e.g. maybe he doesn't want to hurt, maybe he wants to explain. Once the actor gets the motivation a big part of your work is done.
When you have to shoot quickly this type of preparation helps everyone. Screen actors spend their lives surrounded by lights, stands, equipment, props, and fussing people. By the time "action" is called they've been adjusted, touched up, filled, flagged, and cheated to the point of distraction. Getting them centered, in character, and in the moment is your number one priority (otherwise, why are they there). Reminding them of their verbs can put the actors where they have to be, and help you focus on what you're supposed to be getting. Of course there's a lot more to it than this, but this is where I started from.
Two nights before shooting we had a production meeting. I don't remember much about it, except that everyone who had to be there was, and we didn't waste any time. I remember saying thank you, and assuring people that we had CSST (Quebec workplace insurance) coverage, and that safety was absolutely the most important thing. This isn't bullshit -- I've been injured at work, and been on workman's compensation and it pays nothing. Always be safe. The meeting was held at a lawyer's office that Esteban had borrowed for the evening around the corner from my apartment, and it was over fast (people with experience don't hang around). Rene ran things and I left feeling a lot more confident than I had entered.
The afternoon before shooting I went down to the studio and found Jacques rigging the green screen with Danny and Gilles. Walter Diaz (Transport and Art Department assistant) had delivered the props to the second floor studio via forklift. Thank god Walter knew how to operate a forklift (one of those things I hadn't foreseen), and so the small space was crammed to the gills with gear. The green-screen had been freshly painted and unit had provided non-abrasive footwear so everyone could work without destroying the floor - everyone except me; I wear size 13, so I had to walk around in socks.
There were, of course, the usual miscommunications and screw-ups. I won't go into details because everything was fixed with a phone call, or an on-site decision. But there was one classic conundrum which required expert intervention: confusion arose about how best to sync the sound recorder to the camera. I insisted that a cable go into the camera, but Tobias and Jacques also wanted a backup recording (correctly so) in case the wrong button got pressed or a cable malfunctioned. Camera assistants hate having the extra wires running into their machine - it's just another thing to get tripped up on -- but post production people love it because it often saves them a transfer. Tobias and Jacques couldn't seem to figure out how to sync up the HPX170 and the Digital recorder -- what type of time-code? Drop frame or not? Etc., etc.. We all started scanning PDF manuals trying to figure this out when I called my post-production consultant and favourite editor Simon Webb: "Are you using a slate?" he asked. "I guess," I replied. "Then don't worry about it. Just transfer your image and sound to your hard drive, import them into the editor, and sync them up like you would with film." Just like they did in olden times! D'oh!
The day before we started shooting I did some calculations: we were at least ten-thousand dollars over budget. Oops.
Shooting was a blur. Crew call was seven. I left home at five and picked up Guy Bissonnette and Jacques, stopped at Tim Horten's for an Extra Large Coffee, and arrived at the studio for six. Jacques and I went over the story boards. Rene arrived at 6:30, we made any adjustments to the shooting order, and then kicked off the day at seven sharp.
We hit the ground running. I never had to wait for anyone, which was good since we needed to get a shot every fifteen minutes. Jacques drove the technicians on set and Rene kept everyone else camera ready. Vania handled wardrobe and Art Department simultaneously, and Olivier was always on top of hair and makeup and masks. Thanks to Sebastien, Eva, and Esteban my producer duties were now virtually non-existent. The crew was exceptional. They were always ready, always professional, always enthusiastic, always hands-on.
We set up an average shot in five minutes (that's lighting, camera position, and actor blocking), shot it for ten, then moved on to the next set up. How did we do it so fast? Jacques's preparation and the hands-on crew. I'd locked my storyboards several days before shooting. The boards included camera height and focal length, and Jacques had made his own lighting plan. Most of the time he set up the shot without me. Some people consider this level of preparation a hindrance to spontaneity. Nothing could be further from the truth; the boards saved us having to wonder about coverage and let us concentrate on what was happening in front of the camera -- which was the whole reason for being there in the first place. The schedule was six hours shooting, one hour lunch, five hours shooting. I finished up the day doing data transfers, then we went home and started it all again the next morning.
Day one went more or less as planned. Jacques brought out the toys (the most irritating was the "helicopter": two lights on extenders that rotate over the actor to simulate passing street-lights), and I responded by berating him constantly. Months later, in post-production, I was very grateful for that helicopter, as I was for all his lighting gimmicks. In spite of my whining we made the day. Once I got home I'd check the footage quickly, randomly viewing what we shot off one of the portable drives we transferred to during the day (in fact we had three drives to transfer to: my lap-top, and two LaCie Rugged portable drives for backup). Then I'd take my storyboards and cut and paste the frames for the next day into a shooting order.
On day two we lost Matt and Brent early to other commitments. We finished with some inserts of Gillian, and some pistol close-ups, and wrapped -- an hour early! We hadn't made our day, but for some unexplainable reason I was sure we'd pick it up the next day. Besides, there was nothing I could do about it.
Day three was like day one, but with extra scenes! For one shot, Jacques needed significant set-up time so Tobias, Jose, Brent and myself recorded the voice over. We made the day comfortably -- minutes to spare -- wrapped the studio and gear, and I went home happy and exhausted.
There were, of course, highlights: I don't think the crew were ready for my brusque manner with Jacques (I think I was civil with everyone else, and if not I certainly was trying to be); towards the end of day one Jacques and I had a brief but heated argument about some dolly track; the P2 Card reader malfunctioned (replaced by the camera house in 45 minutes with no data loss); Matt played Einstein waking up from a nightmare, and Jacques suggested "he should touch his morning wood," -- we used it; Rene calculated that we did around 150 set-ups in three days (thirty-three shooting hours).
The day after shooting (a Monday) I was on the road at eight, returning a pistol I'd rented from Andrew Campbell (Cold Blue Productions), wrapping up paperwork, filing timesheets. Walter and Vania wrapped the props and costumes, treadmill and misc. stuff and then Walter wrapped the cube truck and closed up the account. I wrote the last checks sent my request to the Canada Council for the balance of the $20,000 grant, and watched my personal bank balance drop to almost zero. Months earlier, before I'd got news about the grant, I booked my annual trip to Australia so everything had to be closed up within the next few days. Wednesday I wrapped up the last petty cash with Vania. Thursday I settled up with ACTRA and Walter. My last conversation was with Esteban on Monday afternoon as I was boarding the plane: he was covering anything outstanding, and I would figure it out when I got back.
We got everything we needed and no one was injured. We were about $12,000 over budget, and my bank balance was almost 0. All in all, things were going better than expected!
I was in Australia from December 3rd 2010 to January 16th 2011 visiting family. I landed in Montreal January 17th and started editing the next day. My original estimate for post had been four months (completion in May), but I told everyone six (completion in July) to cover myself.
My first assemble took about ten days. This might seem excessive for a film that only runs 16 mins, but there was a lot of down time learning the software (Premiere Pro); sorting and logging viewing the material; and looking for tutorials for things which should have been easy to find. Complicating matters was the use of multiple passes to shoot the Twins, and sometimes three passes to shoot Brent and the Twins.
I cut in small scene groups of about 3-4 minutes each, and then nest these inside a larger assembly of the full movie. This lets me completely re-edit a sequence and trying multiple configurations of a sequence without having to alter the larger timeline.
I have trouble cutting for more than four or five hours a day - I lose focus and tend to go blind to the problems in front of me. One trick that helps me spot problems is to watch the material in a mirror. Visual artists have been doing this since the mirror was invented, and it reorients your brain enough that any problems in the image (in this case, edited movie) pop immediately. I also use a technique that Walter Murch describes in his excellent book "In The Blink of An Eye". I run the footage, hit the space bar where I think the cut should be. Then I repeat the procedure and if I get the same frame twice then I'm on to something.
As I mentioned earlier, some of this became quite complicated when I had to layer footage. Matt was playing twins, and I had to sync up the takes (left and right) on different layers in the time line. This was easy when it was just Matt, but several shots involved Fang and the Twins (double the takes and double the set ups. I had to choose which take I would use for (a) Brent, (b) Twin 1, and (c) Twin 2, then sync them up on separate time lines.
Simultaneous to the picture edit was the construction of the 3D backgrounds. I built the models in 3D Studio Max 9 (kindly lent to me by a friend), then unwrapped them and created the texture maps. The models themselves were simple structures, but the texturing required a lot of testing before I arrived at something useful. I used a Wacom Graphics tablet (a cheap one), Photoshop, and Autopainter (an application that turns photographs into paintings) to generate the texture maps. Then I would reapply them to the 3D models, render out the results. This took hours for every model and some of the material still doesn't work properly. Oh well...
Simultaneous to the picture and 3D work was continuing After Effects tutorials. Lots of stuff still had to be learned. The Fang Designing sequence was shot as a single take of Brent walking around a drafting table having great ideas. In my time in Australia I'd come up with a way to make this work, but now I had to generate the material and learn the techniques to do it. That simple montage took over a week (about 60+ hours) and later I re-cut it!
All of the above, however, demonstrates the advantage of a stay home, stand alone system. I could cut at 4AM (and jet lagged I often did) without disturbing anyone; I could literally drop what I was doing and try out an idea. I wasn't paying rent on the post production system, or paying someone to operate it. Just as well, because I went about eight months over budget on my post!
Once the first assemble was completed I focused entirely on the 3D environments and the After Effects integration. Progress was completely unpredictable: sometimes the most complex set-ups dropped into place, and sometimes the simplest ideas took days to sort out. Everything was rendered out into multiple elements (alpha channel, z-buffer, self-illumination, specular, shadow, RGB, etc.) to give me flexibility at the compositing stage. Jacques' lighting consistency was excellent. I matched my 3DS key-light to his, then filled and tweaked until I got a match. Most of the time this was very fast. We shot at 720p, but I rendered the backgrounds at 1080p and resized them. It allowed me to reposition and helped with any aliasing issues that might arise.
I completed most of the 3D sets, then went back to Premiere with a fresh eye for a second assemble edit, refining and fixing as many problems as I could find. The way the movie was structured and shot, there was specific coverage and specific sequence design, so most of this second pass was trimming and altering cutting points. Ultimately I discarded about 5% of the shots we made (about one in twenty). My elaborate montages (the Machine Design and Rooftop sequences) were trimmed to the bare minimum (each of these took days of trial and error before I finally got what I wanted). Because so much of the material was virtual (either 3D or 2D), I needed to be able to move from After Effects to Premiere and back again, to test and re-test ideas. Adobe is NOT a sponsor of this project (although if they would like to be I can be reached at the address below) but the CS5 package (and a lot of third party help) made it possible.
I completed my "semi-final" edit in April. I say "semi-final" because I knew that three sequences were incomplete (the Laboratory, The Forest, and The Last Scene). This was a deliberate omission, and easy to manage because of the way the scenes were nested in the master timeline - I could make alterations up to the last minute without altering the sequences before or after. I stopped editing because the far more daunting task of keying and compositing was about to present itself.
For the uninitiated, keying is where you remove the green background from a shot (you "key out" the background), and compositing is where you replace it ("composite" the foreground on top of something else). There are lots of tutorials and books on the subject (Steve Wright "Digital Compositing for Film and Video" and Marc Christiansen "AE CS5Visual Effects and COmpositing Studio Techniques" are excellent) and if you're interested I suggest you read them. I decided to pull all the keys first, assembly line fashion, and then go back and composite in a second separate pass. I had no experience in keying and everyday quickly turned into an adventure/nightmare. This is where the four-setups-an-hour schedule came back to haunt me. While most of Jacques green-screens keyed effortlessly, any time there was contact between an actor and a surface I had problems. The actors threw shadows, and the shadows desaturated the green. The shadows weren't sharp enough to keep, and not soft enough to key through. Also, the proximity to the green meant that spill (contamination o f the foreground by the background green) corrupted the key. Jacques had been fastidious about masking off any green that might be throwing spill, but there isn't much you can do about an actor lying on the floor wearing a white shirt. Maintaining the grand tradition of post prod supervisors, I cursed Jacques on an hourly basis for things I should've spotted on set. After some very discouraging attempts I hit the books and came up with solutions that worked almost all the time. And when they didn't work, I rotoscoped. This was a situation where a little more experience would have paid off - rotoscoping Freud and Jung, and Fang on the floor was relatively easy, but the tavern sequence with tables, chairs, shadows, reflections, and transparencies (and 3D motion tracking) was a nightmare.
This period was also where I discovered the shortcomings of the HPX170. We had shot everything intending to go to black-and-white (in fact the colors worked so well I was tempted to keep them). The best way to get good skin tone is to convert the colour to black-and-white, while increasing the red balance (thus lightening the skin tones). Since we used a lot of brown in the wardrobe and set pieces, everything shifted. For the skin tones this was fine, but darker tones started to become grainy. The HPX is to a "real" digital camera what 16mm is to 35mm film - and then some. It has a 7:1 compression, and the chip is 4:3 stretched to 16:9 all of which add up to noise (grain). These effects might not be too noticable in close-ups and medium shots, but in wide shot (full figure) the results can be painful -- definitely not ideal for green-screen work. These were the consequences of my pipeline decision, and given the same set of circumstances I would make the same decision again, but it is a cost you have to live with. I also found Keylight 1.2 to be quite grainy.
Jacques' interactive lighting effects made me crazy for a while -- more obligatory cursing.
Problem 1 - we'd used strobe lights in the laboratory scenes to simulate electrical discharges. Rather than try to match the effect in 3DS Max, I rendered out the background animation, including a "normals" pass (look in your help files for a description of "normals") and did a re-light in After Effects. The solution was simply to add a brightness effect to the rendered 3D footage and animate it manually. This took about twenty minutes per shot - about two-or-three hours for the entire sequence. Then I used a piece of software called "normality" (see the software list) to adjust where the light fell. Unfortunately the strobe blew out the green-screen near the floor, making a key impossible, so I was forced to rotoscope feet and chair legs.
Problem 2 - the subtle flickering of gaslight in the tavern. I dreaded having to match this because the frame-to-frame brightness alterations were almost imperceptible. I tried an automated wiggle expression applied to a brightness modifier, and even though it's only the most approximate of matches, it seems to work.
Problem 3 - Jacques' "passing streetlights" effect. This was a two stage solution: first I created a preview animation in 3DS with a grid replacing the floor and lined it up with the background in After Effects. I altered the speed of the background animation (in AE using percentage values) until the actor's feet "stuck" to the 3D floor (or until the passing background looked right). Then I adjusted the 3ds Max camera (using the percentage values from AE), then rendered out the sequence with two or three extra seconds at each end. I imported the 3D sequence into After Effects, positioned it, then "rolled" it back and forth on the timeline until the streetlights in my background animation approximated the lighting effect. This technique really paid off in the Fang Tiring sequence where we dissolve across three different backgrounds moving at three different speeds, with three different lighting cues.
Problem 4 - Fang and the swinging light-bulb, was solved by rendering the background plate, including a "normals" pass and using normality again to relight it in After Effects. I manually matched the After Effects light to the swinging light-bulb in the shot.
Problem 5 - matching the lighting of Woman's Head on Mannequin body was solved on set: we shot Gillian and the Mannequin simultaneously in the same frame, so the lighting effects were the same on each; then I matched up the 3D background the same way as the moving light-bulb.
Eventually I found myself at the last two sequences in the film, both of which I'd avoided fine-cutting until the last possible moment. This was a good move: I looked at the material with fresh eyes, and re-cut both scenes in two mornings (about three hours in total), while using the rest of the day to work on the 3D material and solve some transitional problems: the twisting dolly-in Train Approach, for example, was added when I realised how lifeless the existing shot was; camera shake was added in AE to the 3DS Preview files, and a couple of new shots were invented to keep the final sequence moving. Jacques' big set-up was discarded and replaced with a shot he begged me not to use because of lighting difficulties. The difference in the sequence was night and day - well worth the waiting time. I started the arduous task of setting up and rendering out some prolonged background animation.
A week later I'd finished my first pass at compositing the 3D material. I went back and added muzzle flashes and bullet hits, lightning and electrical flares, and then rendered each sequence out in After Effects, and dropped the final Quicktime files on top of my Premiere Pro timeline. I realized, only as I was doing this, that if there were any sync problems I would find out now. Everything synced up. Of course there were problems: tracking markers that needed to be erased, colour timing problems, etc., etc.. But most of these issues could be resolved with some adjustments in AE and re-rendering.
The idea of "tinting" the footage had been discussed with Jacques at the very beginning. Originally I'd planned on tinting present and past different tones. It seemed like a bit of a cliché, but I couldn't shake the idea that the footage needed something the black and white (which looked very good) wasn't providing. Working on the laboratory 3D model footage I'd done separate outputs for RGB, specular, reflection, self-illumination, etc., when I tried different tints for different layers. Glass was turned green, while everything else (including the live-action material) was tinted brown. I tested this on some other scenes (the green fog in the brown streets, red blood on a brown floor) and decided to run with it. When it came time to add electrical sparks, muzzle flashes, ricochets, etc., I left these elements in their original colour. So, what has evolved is an odd hybrid that people think is colour, but is really a series of monotone layers.
I got my second wind when I started the sound edit. I had been dreading this, because I'd never done it before, but it was the most fun I'd had on the show since that day we previewed the fight sequence in the park. It went smoothly from the very beginning and it was immediately obvious with even the crudest mix that sound effects make the picture look better. Premiere Pro made this stuff very fast, literally dragging and dropping effects into the timeline (gunshots, ricochets), or rolling and trimming them into sync (footsteps loops). The best example was the fight scene, which (after eight months of watching it) I was getting sick of. I used punches, kicks, and the whoosh of swinging arms to punctuate every hit and miss. It took about two hours, and the effect was amazing. But the most useful discovery was using off-camera sound to sell a shot: when Jung tackles Freud and they drop out of frame, it's the sound of them hitting the floor that sells the gag - the sound continues the motion. Ah, the magic of cinema! This kind of thing is second nature to sound editors (or at least should be).
The sound edit took about two weeks of assembling the right noises. I took the advice of an online tutorial and ran off a quick mix onto a CD, which I proceeded to listen to on my stereo, clock radio, and tv speakers. Then I tossed out a bunch of stuff and started again.
The music tracks were just as problematic. I'd worked out some busy material in a Hungarian minor key, and had four or five pieces ready to drop in place. They didn't work at all. By coincidence I'd recently read an old bio of Robert Fripp (an excellent work by Eric Tamm) and found a freeware VSTi plugin that reproduced the "Frippertronics" effect: essentially a long tape delay allowing for the creation of layered, evolving soundscapes. I tried a few things and got what I needed in two or three takes.
I put it all together and ran off a Blu Ray and watched it on a couple of TV monitors - not good. The live-action material was lacking in the blacks, and floating on top of the background. I discovered that I'd been working in an arbitrary color-space, and I'd have to re-correct all my live action material to match the backgrounds. This is too tedious to describe, and often involved redoing keys from scratch. I can't remember how long it took, but it felt like weeks, so it probably was. And I repeated the process several times.
I got back from Australia in late January 2012 and estimated I would wrap 'Fang by end of March. In fact things piled up regularly from this point on: build a web page, look for festivals, get some feedback, design the packaging,.. And I still had to finalise the sound and picture. Having everything on my desktop computer was a huge time saver. Sill, things started to pile up: a spelling mistake in the credits, too much grey in the background, problems burning the DVD/Blu ray files, etc., etc... And finally Jacques convinced me to have another look at the project in black and white. I did, he was right, and we took out all the colour - back in After Effects, shot by shot. Again.
The last pass was a "colour" correction in Premiere Pro: I duplicated the master footage, boosted the contrast so only the white highlights remained, and then blured it and screened it in over the original footage at about 25% opacity. I believe it's now called a diffusion pass, but it's basically a technique Cecil Beaton used when he was printing photographs (Beaton would expose the print, and then make a second exposure at 25% with the image thrown out of focus. Beaton was one of the references in my grant application.
I wrapped the project in April 2012, way behind schedule. In fact I had taken off three months to work in the transport department on a feature film - you know, paying the rent - and eight weeks visiting family in Australia, so I was only about four months behind. The Canada Council sent me a friendly reminder each time I exceeded my deadline. Many many years ago I'd worked on a Canada Council project for a director who spent all the money on his house and car, and I offered to show them my books as an act of good faith. They never accepted the offer, which I think was remarkably trusting (I would've insisted). I logged an average of 60+ hours per week (average) from January 18, 2011, until mid April 2012. Roughly 2,500 hours in post production -- doesn't seem like that much, does it.
At this point in time I am trying to get the movie out there. There's not a lot of market-space for a 17 minute film in black and white, but there's the festival circuit. I have other projects -- film and non-film -- I'm desperate to move on to, and 'Fang has worn out it's welcome.
Was it worth it? Yes.