Blog Posts: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Post-production is a multi-faceted process. Though we now have a good cut, there is much left to do. We've spent most of the last month running a Kickstarter campaign to fund post production. Fortunately it was a success! In the meantime, we've brought on sound editor Ed Callahan who has begun working on sound effects and adjusting the dialogue. The space ship location is a small space, but the stage we were shooting on is large, so the dialogue is "roomy" meaning it has echo and reverb that doesn't sound correct in the setting. Ed will try to manipulate it, but if he can't get it to sound right, we may have to do ADR (automated dialogue recording).
I also did a spotting session with VFX supervisor Tim Hendrix to go through the cut and talk about the effects needed. He seems pretty confident he can do everything required without too much trouble, though he needs to test his 3D workflow. Assistant editor Matt Kendrick has been setting up files for Tim to work with and for stereo correction. We are now trying to find time for James to do his stereo correction work, hopefully before Thanksgiving.
Thanks to all who contributed to our Kickstarter campaign - without you we wouldn't be able to finish the film!
October 6, 2014
We've been editing in 2D. Now that we've got a cut we're happy with, it was time to look at the film in 3D (or "stereo" as the pros like to call it) with James, our stereographer, to make sure the cut will work without undue eyestrain. We tried to do this viewing yesterday, but it turns out the edit had lost the stereo information when the temp effects were added, so Matt, the Assistant Editor, had to re-do the cut with the original shots (and without temp effects, which was fine for our purposes). So we reconvened today and saw the film in 3D. It looked great!!!
There were a few issues, but James assures me they can be fixed in post processing with a piece of software called Pure that can adjust the interoccular distance (distance between camera lenses) and the convergence (how far from the camera equates to screen plane). Doing this causes thin black bars at the edge of frame, so we may have to push in digitally on some of the shots, but only slightly. There is one shot that will be problematic - a point-of-view shot of Pike approaching Foley floating in fetal position. Pike's hand comes into frame too close to camera "breaking" the 3D (in essence, the images in the left and right eye are too far apart for your eyes to bring them into alignment). We can adjust this with Pure, but as the hand gets fixed, it starts to break Foley and the background. We will futz with this and see what we can do, but we may have to let the hand break. The focus of the action is Foley and I really don't want to lose the shot. It'll be a little like having an out-of-focus object in the foreground of a 2D film.
Part of this was for me to learn, and this shot has definitely taught me something!
October 1, 2014
Editor Howard Heard and I have been going back and forth with various rough cuts. We've reached the point where we're fairly happy with the cut, so today we showed it to a couple of filmmakers (Mark and Marilyn) to get a fresh perspective and make sure the story beats were clear. Thankfully, both of our viewers followed every beat! The major feedback was related to tone, that it was unclear until the moment of the bite. We will adjust some of the edits to try to make the performances more serious, but I believe this issue will mostly be helped when we have score (currently the cut is without music) and when we do color timing (Andrew and Mario lit the film brightly to give the VFX team more picture information to work with, but the intention was always to adjust to a darker, more moodier look when we color time the film).
September 14, 2014
We're into post-production! The last week was spent with Matt, our DIT and Assistant Editor (and sound guy), syncing the sound on all the dailies and getting the files set up for Howard, our editor. The rest of us caught up on some much needed rest! Today, Howard and I went through all the footage and I gave my notes on which takes I liked and why, and how I saw certain scenes cutting together. Next Howard will do a rough cut of the film. Then Tim will start working on the effects while we work on sound and score. Last will come color timing and sound mixing. And somewhere in there toward the end we'll do some tweaking of the 3D convergence.
Watching dailies always involves moments of excitement when things look better than I feared, and moments of disappointment when I realize I didn't quite get what I was looking for in a take. But thankfully the Microbe dailies mostly looked fantastic. There is nothing I am worried about cutting together, which is rare. Of course we may have some unpleasant discoveries in editing, and then again when we see the visual effects, but I have footage I like covering pretty much every moment of the film. It's a challenge on set to be sure everything comes together in a take - the performance, camera, sound, props, and various practical effects (we had several scenes with props on wires). The difficulties were compounded for us by the need for actors to move as if in zero gravity and by the 3D. My primary focus on set was watching the performance. I have to rely on other people to alert me if there are problems in their departments. One thing I will pay more attention to in the future is checking with each department to be sure they think the take was great before we move on. But one of the beauties of editing is not every take has to be perfect to be usable. For example, we can use half of one take and half of another if we're going to cut to a different camera angle in between. So, for the time being, I am pretty confident we will have an awesome film!
September 8, 2014
We finished production yesterday! What a great feeling!!! Yesterday was a little more rocky than the first two days and we went an hour and a half over. Thankfully we had planned a very manageable schedule for the last day in case of problems. Even so, we had to cut a couple of "wish list" shots, and we forgot to do an insert which, fortunately, we can live without. The wire work was difficult, as expected. One particular difficulty was trying to hide the vests under the costumes. We had picked dark colors parly for this purpose, and most of the wire work will be quick action, but even so Tim, our VFX supervisor, will probably have to do more work than he'd like to clean it up. The biggest problem was at one point the "travelling rig" - a support system that moves horizontally along the I-beam in the ceiling, tore part of the ceiling material and a cube of fiber glass insulation fell down, forcing us to halt work for safety reasons until it could be repaired. That set us back and ultimately accounted for the overtime. But, that's production - things go wrong. And because we had a fantastic crew, everyone pitched in and made sure we got the shoot done. I'm very happy that the stunts were all done with utmost safety.
And just a word about that crew - somehow we were incredibly fortunate. Every single person had a great attitude. Even when things got tense it was still a positive, upbeat team. One thing that struck me was the ballet of production when you have a crew of people who know their jobs. When the AD (Che) called "moving on," indicating a new shot, art department (led by Reed) would fly walls in or out as needed while the gaffer (Mario) set up lights while the grip (Vicky) moved C-stands while the camera department (led by Andy) laid dolly track... somehow never getting in each other's way (or almost never). If you stopped to consider how much was happening very fast with such precise coordination it was stunning.
At the end of the day we had gotten everything we needed and there were good spirits all around. Now it's on to post-production. Can't wait to see how it turns out!
September 6, 2014
Day two was supposed to be our shortest day. Didn't quite work out that way, but we finished on time and got everything we need. We got some great footage of Pike (Scott Peat) hiding in the locker while Foley (Joan Farrell) tries to get in to kill him - with the only light through slats in the locker vent! We also did the opening scene where we establish the zero gravity with a really cool looking floating wrench. The actors are amazingly talented at faking zero-G movements. For many of the close shots we put them on big exercise balls so they can "float" realistically, and have a less rigid posture. We also took a few moments for some pictures inside the completed set. The crew is great - everyone is easy going, hard working and good natured. It's been a complete blast so far. Tomorrow we do wire work. I'm nervous about that, never having done it before, but we have a very experienced stunt coordinator in Kimo Keoke.
September 5, 2014
First day of production! Everything went fairly smoothly. We got all the shots we planned and finished just ahead of schedule. Today had the greatest number of shots, but most were fairly easy so we could ease into the shoot and figure out how to work with the set, which we reconfigure to represent different locations in the spaceship. Also, many of us were learning on the go about shooting 3D (James Tucker-Robbins is our experienced stereographer helping us out!). We have a huge 3D monitor to see what the shots look like. The crew seems to enjoy passing around the glasses to check out the effect. Really the entire cast and crew did a fantastic, professional job, and it seemed like everyone had fun. We discovered Amy Dallen, who plays Madison, has a great talent of balancing on one leg and moving sideways into a shot. It really looks like she's floating, and it's all her! Scott Peat (Pike) and Joan Farrell (Foley) did some excellent stage fighting. Joan was incredibly scary playing "infected." Probably the biggest issue was the fake bite wound on Scott's arm which could easily get smudged or damaged... but that was really not a huge challenge.
September 4, 2014
Today was prep day for the shoot - which means tomorrow we start production! Today Reed and his team moved the set to the sound stage. Kimo set up the rigging for the wire work. And Andrew and his team picked up and prepped the camera package. Meanwhile, the rest of the team dealt with last minute planning, shot lists, etc. So far one of the biggest challenges has actually been wardrobe. I probably didn't give enough attention to that earlier, and as a result we had a little difficulty getting everything we need. Not that the wardrobe is complicated - it's all off-the-shelf. But we need several of each piece in the required size, and it turns out a lot of stores only carry one or two pieces of each article of clothing in each size. But we finally got what we needed - we hope! The other concern is hiding the wire rig harnesses under the costumes, which need to be fairly fitted so as to not compromise the zero G effects. Crossing my fingers on that one. Call sheets just went out... looking forward to tomorrow. Very exciting!
September 2, 2014
Today we had a pre-production meeting with our key department heads: producer Roxy Shih, assistant director Che Thornhill, cinematographer; Andrew Shuford, production designer; Reed Johns, stunt coordinator Kimo Keoke, and VFX supervisor Tim Hendrix. It was intense! This is a complicated production and there is lots to work out. But the team is good and I think we have it all under control. A big part of the meeting was ensuring that we know exactly what we are shooting each day, and that we get every shot we need to complete the film. Andrew and I also visited the workshop where the set is being built. Here are some pictures:
August 28, 2014
So we're now getting into the nitty-gritty of pre-production which means people are actually buying things. And often I'm discovering that the estimated prices don't match actual costs. Roxy has been doing a great job keeping the total budget within an agreed upon range, but it's meant some creative compromises. Well, that's filmmaking! The first big added expense came when we found the soundstage we had planned on was unsuitable for our needs, both in terms of size and what we needed for the wire work stunts. So we switched to the fantastic Greenery Studios, who worked out a deal for us, but still cost a couple thousand more than budgeted. The next issue was the production design. I kind of always knew the budget for the set was probably too low, but when the numbers started coming in it threatened to derail the whole project. So Reed and I put our heads together to redesign the set in a form that is more affordable... and Roxy found some additional money in the budget. I will have to rethink some of my shots, but it shouldn't compromise the story at all. Filmmaking = problem solving! Gotta stay flexible.
We are crewing up and have some great people on board in all kinds of positions. I'm very grateful for everyone's help - everybody is taking reduced rates or volunteering time to make this film happen. It's going to be a real team project. Now I gotta go redraw some storyboards!
August 21, 2014
As the various departments are starting to come together with their plans and designs, I've created the master shot list. In addition to typical the typical scene-shot breakdown, mine also breaks down the elements - background plates, CG, etc. I then divided the shot/element list across the three days, grouping them by location within the spaceship so we can redress the set after each day's shots are complete and not hold up shooting. I've also got to be cognizant of when Joan will be in FX make-up and when she'll be in normal make-up so we don't have long delays while the FX make-up is applied. We'll start with all the green screen elements on Friday morning, then build from there.
Less than three weeks until we start shooting!
August 19, 2014
For various reasons, Rachelle had to drop out as our production designer, but we have a fantastic new production designer, Reed Johns! Met with him this morning to start going over the set design, discussing how we can use a standard architecture redressed for the various areas of the spaceship. We also discussed such technical elements as flyaway walls and built-in lighting. At the same time I met Kimo, our stunt coordinator. We discussed the wire work needed and what that would require. Much of the wire work will be done against greenscreen, of course, but we're debating if any is going to happen within the sets, which would require the sets to be built to accomodate the wires. Regardless, we want to do all the wire work on the first day so we can wrap the stunt crew out after that. The second two days will be devoted to close and medium shots where the zero G effect will be created by balancing the actors on exercise balls.
July 10, 2014
We have finally got our final cast member - Amy Dallen will be playing Madison. We had a table read of the script today so Amy could meet Scott and Joan and we could talk about logistics. Amy has a great "little sister" vibe with Scott and Joan that will play well for the film. We also recorded the audio from the reading so Tim can use it in his "ghetto pre-vis" film.
June 1st, 2014
Big pre-vis (pre-visualization) meeting with the producer (Roxy), director of photography (Andy), editor (Howard), production designer (Rachelle) and VFX supervisor (Tim). The goal was to go through the shot list Andy and I made and figure out how we will achieve each shot. Tim is encouraging us to do as many of the effects practically in-camera (which I like) and also to mix up the way we achieve the microgravity look. He says that varying the way we achieve an effect will keep the audience from figuring out our methodology and help sell the look. We are somewhat concerned that the actors will have to be in good shape for what we're going to ask them to do, and look effortless while they do it! We're also keeping a close eye on safety.
There were several times in the meeting when Andy and I presented a shot and people went, "oh, cool!" So this is encouraging me that this could be a really great project. But it's a challenge. It's kind of a binary proposition - either we pull off the effect, in which case I think it will look awesome, or we don't, in which case I think it could look horribly silly.
May 22, 2014
Found my new production designer, Rachelle Flauta! We met to discuss my conceptions for the design of the set and compared some reference art ideas. My idea is to use a basic tubular unit as the underlying structure of the space ship. This is inspired by the International Space Station (ISS) which is composed of units brought up by the space shuttle, so each unit is the size of the shuttle's cargo bay. If we build two such units with flyaway walls, we can assemble them in various configurations and redress them to represent all of the areas needed for the film. We will likely have to composite in set extensions in post for a few shots. A lot of the design will reflect the microgravity environment. For example, in microgravity there is no up or down, so computers, storage, etc. can be on all of the surfaces. Also, in microgravity there can be no large rooms - otherwise the astronauts might get "stuck" in the middle of the room. They always have to be arm's length from a wall to push off of. We will make our set units 8' diameter as a result (this aspect is what encouraged me we could build the set at a reasonable budget). Another interesting aspect we will steal from the ISS: handles and straps along all surfaces for astronauts to pull themselves along and slip their feet into to hold themselves in place. I want the look to be realistic for a spaceship a couple decades in the future - no retro touches, but also no "grit."
May 15, 2014
Met with Andy today to go through the script scene by scene and break down our preliminary shot list. Due to the technical nature of the project, we have to have very precise planning. Creating the shot list requires analyzing the best way to get the story and emotion across on screen, but also requires balancing the reality of budget and shooting schedule. We have to pick the right places to do the most complex, visually exciting shots, places where they will have the maximum payoff. When we're doing a dialog scene, we keep the shots simple. Also key to this process is figuring out exactly how we are going to physically accomplish each shot. We made a list of questions for Tim about how best to do the various effects. Andy also wants to build a lot of the lighting into the set, so that will require coordination with the production designer
April 30, 2014
Met today with a producer, Roxy Shih, who is coming on board the project. She was recommended by my new VFX supervisor, Tim Hendrix, a former student of mine who just graduated. Very excited to have both of them. Roxy is full of energy and a get-er-done type. I definitely need that help to manage all the logistic and organizational details so I can really start focusing on the creative. Roxy's first task will be finding me a production designer. It's great to have Tim's participation because he knows all the latest VFX tech stuff, but he has a preference for doing as much in-camera and practically as possible. I have a similar philosophy. The project is starting to ramp up again. We're setting a target shoot date of the end of June, depending on stage availability, etc.
Ian has unfortunately had to leave the project due to competing work. Fortunately, my friend Andy Shuford has come on board as my new Director of Photography. However this transition, and more financing problems, has forced us to push back shooting into next year. Stay tuned.
August 21st, 2013
After we lost our initial camera package, the project was in some trouble. We didn't have enough money to finish it the way we planned. So for the last several months, I and the rest of the Microbe team have been looking for alternative sources of financing or discounts on equipment that would allow us to do the project with the funding we have.
Director of Photography Ian McGlocklin suggested we look into converting the film to 3D in post rather than shooting on set. This would result in considerable production budget savings. With a stereographic (3d) camera set-up, we would need two cameras as well as a beam splitting rig. Shooting in 2D would allow us to eliminate one camera as well as the beam splitter. Shooting stereo also requires a stereographer, so we could eliminate one crew position. But perhaps most enticing, shooting with a stereo rig takes longer than traditional shooting. The cameras must be constantly readjusted to get the proper alignment between them. Adding stereo in post might possibly allow us to reduce the length of the shoot from three days to two. This would mean one less day of renting stage space and equipment and one less day of craft service and catering.
So what's the downside? Post stereo conversion is done by using a computer to separate elements within a shot and put them at different depth levels. There is cheap software to do this on your home computer. However, the result is not great with this lower end software. Some of it doesn't really identify individual elements, it just applies different depth to different parts of the frame. That is not what we were looking for when we decided to do Microbe in 3d. So, we would have to go to a professional post house. Post conversion 3d takes time and that means money. We might have an advantage in that we are shooting many of our elements separately and compositing them together, so we wouldn't need to rotoscope everything to assign it depth. In essence, many of our elements would be pre-rotoscoped. However talking to post houses, their initial response was that it would be cheaper just to shoot 3d on set. (Interestingly, most of the DP I talked to suggested post-conversion is easier and cheaper while most post-conversion people suggest shooting native stereo is easier and cheaper. It seems 3d is a pain wherever you do it.)
One of my students suggested a third approach: shooting native 3d on one day only for the most dynamic shots, then shooting close-ups and inserts on the remaining days and doing post-conversion on those comparatively easier shots. This is still an option we are looking into.
We will be making this decision soon. We have to... we have now committed to a shoot date of October 11-13!
The design of the spaceship set in Microbe is going to be different than any spaceship we've seen on film before. Structurally, I was inspired by the actual design of the International Space Station. That consists of several long passages that cross each other. This is partly because most of the components were brought up in the shuttle and had to fit in its cargo bay. But the design is also key for operating in microgravity.
One of the considerations of a microgravity environment is that an astronaut needs to be able to push off of something to move. If you build large, open rooms, then the astronaut could theoretically get stuck in the middle where he can't reach the walls. So a passage style design makes sense - there's always a wall within reach. This is great for 3D because we get long Z-axis (the direction away from camera into depth) when we shoot down the passages. And from a dramatic perspective, crossing passages give us nice place for people to hide.
Recently, the Chinese put astronauts in their test space station. One thing I noticed from the pictures was that the walls of the Chinese station are padded. This makes sense. If people are floating around, it would be easy for them to bump their heads. So I'm incorporating that into my design for the Microbe ship.
I've also thought about the general look of the interior. A manned mission to Mars would require the longest time in space any person has attempted. There would need to be some thought to making the ship feel livable for the astronauts' psychological health. So, perhaps the walls would be light green instead of white. And maybe there would be some attempt to hide the wires and pipes. So that's part of my design concept for Microbe as well.
Of course I've also had to think about practical shooting considerations, but there my design actually gives me an advantage. If I assume the ship is constructed of identical component sections like the International Space Station, then I can build a single section and simply re-dress it to play different parts of the ship. Two of the walls can be flyaway for ease of shooting. And I can extend the length of the passage by shooting plates of the set to composite behind the action in post.
It was actually this design that convinced me I could do Microbe at a feasible budget.
May 7, 2012
Had lunch with a gentleman who works at 3D company 3ality and learned a few more things about shooting well in 3D. First, I learned the downside of a side-by-side camera is that it can narrow the depth range.
A quick primer on some 3D cinematography terminology may be necessary here. The distance between the camera lenses is called INTERAXIAL SEPARATION. The spatial difference between where an object is place in the two frames is called PARALAX - that's what makes the image appear three-dimensional. SCREEN PLANE or STEREO WINDOW is the plane on the set where an object will appear to exist right at the screen to a viewer. Objects that appear in front of the screen have POSITIVE PARALLAX while objects that appear behind the screen have NEGATIVE PARALLAX.
The problem is that when a side-by-side camera focuses on an object close to the lens, an object in the far distance can actually have so much parallax that it goes out of alignment and will look weird and cause headaches. So if I want to have much depth in the shot, the screen plane will have to be 10-15 feet from the camera. And that means staging most of the action 10-15 feet away from camera.
This may not be a deal breaker. But since I plan to shoot physical rear plates to extend the set in post, I may at some point need to move the camera more than 30 feet from the set... and that will require a fairly large stage.
There are a few other things I've been learning about the more creative aspects of shooting 3D:
Changing the distance the screen plane is from the camera is difficult on the eyes. Quick cuts with drastic changes can cause headaches. Good stereographers will sometimes start changing the screen plane six frames or so before a cut so that the transition goes a little more smoothly. (This is apparently done in post - the synchronization of the two images is changed slightly leading in and out of the cut.)
Also, good stereographers will make a 3D plan for the shoot. Perhaps counter-intuitively, I'm told it's generally best to bring the most intimate, emotional moments forward while letting action play out deeper in negative parallax. I will have to think carefully about these things when planning my shoot.
My biggest takeaways from this lunch were the need for an educated stereographer and the need for careful planning. But I was going to have both of those anyway!
April 25, 2012
I've been meeting with various people who know about such things to figure out what equipment I need for the short. My first focus was on the 3D camera.
There are two possible camera configurations for 3D. The first is a beam splitter rig. This is a device that takes in the image and splits it using a half-silvered mirror into two separate cameras. The advantages are the ability to use any camera type with the rig and greater range to the 3D. The disadvantages are that it is fairly expensive and requires more time to set up. It also tends to get out of alignment over time with moving shots.
The second option is a side-by-side camera. This is a camera that has two lenses to capture the image in a single body. The advantage is ease of use and cost, the disadvantage is that it doesn't generally provide quite as dramatic 3D. Or rather, you have to make certain that you are getting greater 3D out of it.
I decided to go with the second option for reasons of budget and because this is my first attempt at 3D, so I wanted to make it easier on myself. I will have a stereographer on set to make sure that I am getting the maximum effect from the camera.
Years ago I looked into doing green screen with mini-DV and discovered that it wouldn't work because of the color compression. In order to do green screen, you need to be able to precisely identify the green color when you composite. But that requires very detailed color information in your digital format. Mini-DV does 4:0:0 color compression, but to get a clean composite, you need 4:2:2. (I don't really know what the numbers mean, but I've seen the poor results with the wrong compression!)
We'll be working at much higher resolution than mini-DV, but higher resolution formats don't necessarily mean deeper color resolution. Several cameras I looked at did not have the capability. The one that did is the Panasonic AG-3DA1. It records in MPEG-4, which can be set for 4:2:2 color compression (and I will have to be certain we set it properly!) Fortunately it also happens to be a relatively inexpensive camera to rent!
So the first decision has been made. And I'm glad I knew about the color compression... I can only imagine how bummed I'd be if I shot with a different camera and discovered the composites looked awful.