Nofilmschool today posted a brilliant short film called The Secret World of Foley, which inspired me to share this DVD extra from Stop/Eject. It looks at the whole sound process on this short film, from production audio, through ADR, foley, effects and music to the final mix.
And here’s The Secret World of Foley – well worth a watch.
Back in November 2012 I posted and analysed the preproduction and production budget for my short film Stop/Eject, a 17 minute fantasy-drama which was shot on a DSLR. Now I’m going to do the same for the postproduction budget, including distribution and marketing. We’re currently selling Blu-rays and digital rentals of Stop/Eject, to raise money for further film festival entries, so please support us by buying a copy if you find this blog useful.
Download the budget here as a PDF (35kb).As you can see, we had almost £2,000 available to us in post, some of which was left over from production, but most of which came from a crowd-funding campaign. You can read my evaluation of that campaign in an earlier post.
None of the cast or crew were paid at any stage of making Stop/Eject, and indeed I tried not to spend anything at all on postproduction. The VFX artists worked on their home computers, editor Miguel Ferros used his own Mac-based Avid system, colourist Michael Stirling used his company‘s DaVinci Resolve projection grading suite, sound editor and designer Henning Knoepfel used his own Mac for the audio work and pulled in a favour to get a free studio day for the ADR, and re-recording mixer Jose Pereira used the studio at the college where he lectures. Scott Benzie composed the music in his home studio and we recorded it for free with four live players at Worcester Tech College.
So except for a suite of clock sound effects, which Henning convinced me were necessary to help the audio tell the story, the main costs in postproduction were those incurred by people travelling so that we could be in the same room for some of the work, and eating lunch on those occasions. It’s important to at least make sure people are fed when you can’t pay them a fee. The most expensive of these days was the ADR session, which involved me and two lead actors travelling from Hereford, Birmingham and Bath respectively to the studio in east London.
Even in today’s digital world, some files are just too damn big to send online, and such was the case with the Avid media output by Miguel ready for the grade. I therefore purchased a USB hard drive, which ended up being couriered across London a couple of times to get to where it needed to be. After the film was completed, I used the same drive to archive all of the Stop/Eject assets and project files.
In order to run the crowd-funding campaign, which lasted for most of postproduction, we needed to build our own website and cut a trailer using library music. We also attended several events to promote the campaign and the film in general, one of which charged an entry fee.
The £79.47 spent on producing the crowd-funding rewards (a.k.a. perks or gifts) was racked up mostly by the hardback glossy script books, costing about £25 each (ex. VAT). The sponsorship level required to qualify for one of these books was £100, and since you also got a DVD, Blu-ray and premiere invite for that amount, there can scarcely have been £60 left of the donation for us to spend on actually making the film! It just goes to show that you should carefully cost up your rewards before you offer them.
That £79.47 isn’t the whole story though, since the next three items listed – the screening venue hire, Blu-ray stock and dupes – were all partly for sponsors as well. (A £10 donation got you an invite to the premiere, £30 got you a DVD and an invite, and £50 got you a Blu-ray, a DVD and an invite.)
£25.61 bought me fourteen blank Blu-ray discs, most of which I got through in trial and error as I authored and tested my first ever BD. The £265.30 spent on dupes got us 60 DVDs and 50 BDs, all with full colour on-disc artwork, inlays and cases. 20 of those discs went to sponsors and approximately 50 to cast and crew, with the rest being reserved for press and festivals.
I deliberately completed the discs in time for the premiere so that I could hand many of them out in person and reduce postage costs. For those that I did post, I used only pre-loved jiffy bags which I had been collecting for some time.
All in all, I’d say almost 11% of the £1,584 raised through crowd-funding was spent on creating and delivering rewards, a little more than I would have liked. Ideally you want to spend no more than 10% of your budget on rewards.
As detailed in another post, I created Stop/Eject’s Digital Cinema Package at home using free software, but did have to buy a hard drive for it and a flight case, since I intended to ship it to international festivals for screenings.
To promote the film at festivals and beyond, we had 50 full colour folders printed, each containing five single-sided monochrome pages of text. We also paid £10 to submit Stop/Eject to The London Film Review, the hope being that good reviews would increase our chances of festival selection.
And that only really leaves the festival entry fees themselves. We’ve entered 25 to date, and the money we’re raising now should allow us to enter another 20 or so. In a future post I’ll provide a list of the festivals entered, their deadlines and fees, and the selection results.
Following my personal observations on the shot planning process the other day, here’s a look at that process in action and a record of some of the thoughts that go through my head as a director when I’m choosing camera angles.
Everything begins with the script, and here is the extract for the Stop/Eject sequence I’m going to break down:
14. INT. ALCOVE/EXT. RIVER GARDENS – DAY – INTERCUTKATE stands behind the alcove’s curtain with an armful of tapes. She pushes one into the recorder – “JULY 16th 2007, 5-6:30pm” - and hits PLAY. Warm summer sunshine steals in through the crack in the curtain. She pulls it back to reveal the river, sunlight dancing and sparkling in the water of the weir.COPY-KATE cycles through the gardens on a creaky old bicycle with a custom paint job and various doodads hanging off, oblivious to her other self and the alcove stood in the middle of a Victorian bandstand.Copy-Kate spots a strange figure on the riverbank, wearing closed-back headphones and waving a big, fluffy microphone at the running water. She looks ahead – she’s about to run over TWO YOUNG GIRLS. She grips the brakes tightly and the bike screeches to a stop with a noise like a small army of warring cats. She catches her breath as the older girl scowls and drags her sister away.
Sophie drew the following storyboards for this sequence, based on my rough sketches:
I don’t like starting scenes with establishing shots; I prefer to reveal them gradually. So when I conceived the first shot (top left) – setting up Kate in the alcove in the shop – I suspected I would probably end up cutting it, and sure enough I never even filmed it. The audience would know by now where the tape recorder alcove was, I figured.
The next shot (top right) follows Kate as she puts down the stack of tapes. This is fairly basic visual storytelling. The audience already knows that the tapes contain recordings of Kate’s life. When we see her come in with an armful of cassettes, we anticipate her nostalgia trip.
As this was the first time Kate was to travel back in time more than a few hours, I felt it important to show the action of the tape going into the recorder in close-up (bottom), to ensure the audience understood the connection between the tapes, the machine and the time travelling.
We then return (top left) to the previous angle, following Kate as she stands back up and opens the curtain. One of my regrets with Stop/Eject was that I never shot over Kate’s shoulder as she looked out of the alcove. I can only think this is because I was trying to avoid doing “the obvious thing”. In this scene I chose instead to tease what she’s seeing, revealing first the sunlight on her face, and then (top right) an abstract close-up of a spinning bike wheel, part of the visual theme of circles I had smart-arsedly developed for the film. My thinking was that time travel was a big and unbelievable concept for Kate to take in, so it needed to be broken to her (and therefore us) gradually.
Finally the scene is revealed (bottom left) in a high wide shot to establish the geography, which then cranes down to draw us into the action. On the day, there was a bush in the foreground, which began to obscure the action as we craned down, so we decided to crane up instead, rising up over the bush to reveal the action.
Next it was necessary to show the place of Kate and the alcove in the geography. I wanted to echo the formality and symmetry of the bandstand’s architecture by framing it flat-on, dead centre (bottom right).
Then Copy-Kate sees Dan, her future husband, for the very first time. I wanted to show an immediate connection using an over-the-shoulder shot-reverse. Since Copy-Kate was on a moving bike, this meant panning with her for her angle (top) and then tracking with her for Dan’s angle (bottom) in order to keep her shoulder in frame. I left Dan’s shoulder out of Kate’s shot since he hasn’t seen her yet and so hasn’t made a connection.
The editing podcast below from summer 2012 explains the various iterations I went through with this sequence. (I later brought Miguel Ferros on board to re-edit the film, and his final version is far superior to all of my attempts.) You can see in the podcast some of the problems that my linear shot planning approach caused, notably my failure to cover the whole scene in the crane shot, and the restrictions which that placed on me in the edit.
Despite these minor quibbles, I’m very proud of Stop/Eject and its visual storytelling. It’s recently received a couple of glowing reviews on Unsung Films and The London Film Review, the latter praising its visuals, and both quite rightly lauding Georgina Sherrington’s brilliant lead performance.
One thing I often find myself struggling with as a filmmaker is clarity of motivation and storyline. It’s amazing how easily an audience can misinterpret something – or perhaps I should say how easily they can interpret it differently from the director, writer, etc. Here are some examples:
Audiences aren’t stupid; you just have to remember that they haven’t read the script, been on the set and worked on the edit for months. They’re coming to it completely fresh, and if the right clues aren’t in the film, they have little chance of interpreting it as you intended.
This is why test screenings are so important. Happily most of these types of issues can be resolved fairly easily by cutting something out or adding a line of ADR, but unless you show your edit to fresh eyes you probably won’t even know they are issues in the first place.
Last weekend Stop/Eject, two years in the making, got its first proper screening at the Mac in Birmingham, to an invited audience of cast, crew and sponsors.
The film had been in postproduction for fifteen months – as long as my feature Soul Searcher. With every film I make I want the quality to improve, which means more time, and I want to do fewer jobs myself and turn them over to talented specialists, which again means more time because they’re all fitting it in as a favour around paying work.
So it might have taken a while, but it was worth it, and it seems that everyone who came along on Saturday agreed. Those of you cast/crew/sponsors who have a Blu-ray copy will be able to see the huge difference wrought by editor Miguel Ferros versus my original cut.
But even since picture lock back in January the film has come on leaps and bounds. The actors were all convinced that the edit had changed since they saw it at the ADR session, but actually all that had changed was the soundtrack. Henning Knoepfel’s sound design and Scott Benzie’s beautiful music, both delicately mixed by Jose Pereira, bring a whole new dimension to the film. Again, if you have the DVD or Blu-ray, be sure to check out the Superior Sound Reproduction featurette for a glimpse into the transformative process of postproduction audio.
The day itself was really nice, with most of the key people attending and everyone having a good time. After the screening many of us went on to Fletcher’s Bar and Restaurant to drink, chat and gather around a TV to watch Record & Play: The Making of Stop/Eject.
As usual with these events there is a tinge of sadness. Will I ever see some of these lovely people again? Will we get to work together in the future? With all the good will going round at a premiere, you want to start shooting a new film with the same team the next day, but of course it never works out that way.
Anyway, thanks once again to everyone who came and everyone who supported Stop/Eject. Venice Film Festival has already turned down the film, but there are plenty more to enter and you’ll all get to see it eventually on the festival circuit or (ultimately) online.
Here’s a VFX breakdown that didn’t make it onto the Stop/Eject DVD, showing how rotoscoping and compositing techniques were brought to bear on a shot of Kate passing through Dan.
Now that huge reels of 35mm film are all but obsolete, Digital Cinema Packages (DCPs) are the new means of getting a film to a cinema. Many top film festivals will only screen off a DCP or 35mm print, and in terms of picture and sound quality and compatibility it is your best option for screening at theatrical venues in general. Much has been written about how you can make a DCP at home for nothing, but having just gone through the process myself for Stop/Eject I’m going to round up some of the best sources of information I came across and enlarge on the area of disc formatting which hasn’t been too well covered elsewhere.
To ensure maximum compatibility of your DCP you need to:
I’ve spoken to filmmakers who have ignored many or all of the above and still run their DCPs successfully in cinemas, but I decided to play it safe and do all of the above, except the Cru Dataport, which was a little too expensive. Instead I bought a 500GB LaCie Rugged USB drive and put my DCP on there. Read Knut Erik Evensen’s excellent blog post on DCP delivery for more info on Dataports and USB compatibility.
So, here is the process I went through. I was starting out with 25,409 uncompressed 16-bit TIFF files representing each individual frame of Stop/Eject, and six mono 24-bit linear PCM WAV files for the 5.1 surround soundtrack. The TIFFs were in 1080P (1920×1080) letterboxed to an aspect ratio of 2.35:1.
After completing the DCP I took it to the Courtyard, my local arts centre, where head projectionist Simon Nicholls was kind enough to let me test it. To my very pleasant surprise it worked perfectly, uploading at about real time via the Doremi server’s USB 2 socket and playing shortly afterwards with superb sound and picture quality. Much as I love celluloid, the ease and cheapness of this process are breathtaking, the purchase of the hard drive being the only cost. I’ll let you know how I get on running it at other cinemas.
Stop/Eject is my fourth major project to include visual effects, and also the fourth where it’s been a struggle to get all the visual effects done. As any micro-budget filmmaker knows, it’s par for the course for some cast and crew to pull out, sometimes without warning or explanation, and VFX artists are no exception. On Soul Searcher, for example, I needed a CG artist for the 80+ shots featuring “spectral umbilical cords”. Four artists started the work and then quit, citing various excuses from exploding PCs to miscarriages, before the fifth delivered the goods.
Stop/Eject has a surprising 31 VFX shots (most of which you’d never know were VFX shots), of which the twelve simplest were handled by me and Miguel, the editor. With the remaining nineteen needing to be outsourced, how did I apply what I’d learnt from my previous projects?
I want to say a huge thanks to those artists who came through for the project: David Robinson, Mary Lapena, Matt Collett, Eranga Mudiyanselage, Dominic Stephenson and Naveed Aftab. You all worked incredibly hard and produced fantastic results – you should be proud of yourselves.
Finally, a few technical points about our workflow, for anyone interested in such things. We shot on a DSLR, so the source footage was in H.264, a format that due to its structure cannot be trimmed without losing a generation. So I supplied the VFX artists with the entire take (along with details of the in and out timecodes of the piece used in the edit) and asked them to deliver their finished shots as 16-bit TIFF sequences. This ensured that we would lose zero quality. The downside to this workflow is that there is a danger of errors being made with the timecodes, leading to a shot not being long enough when you go to conform the edit…. Yes, that happened. There must be a better way. What’s your workflow for DSLR projects with VFX?
I can at last reveal the full list of bonus features that will grace the DVD and Blu-ray editions of Stop/Eject. Please note these discs are for sponsors, cast and crew only. It is likely to be 18 months or so before the film is available to the general public, other than at festivals.
On Monday, Stop/Eject entered the final phase of postproduction: grading. In the optical days, this process involved adjusting the cocktail of chemicals and the length of time the film would be bathed in those chemicals to make basic adjustments to the amount of red, green and blue in each shot and the brightness.
The reason for this can be most readily appreciated if you imagine a scene shot outdoors, in which one camera angle may have been recorded under warm, direct sunlight, whereas another which is cut to immediately after may have been recorded when the sun was behind a cloud and the light was cooler in colour. But even artificially lit scenes will need a little work to match up each shot to the next; the human eye is quite sensitive to changes in colour and brightness.
With the digital revolution came an exponentially expanded toolset for grading. Individual colours can be isolated and changed, shadows and highlights can be adjusted independently, and feathered masks can be applied to highlight or shade just one part of the frame – with the software even able to track elements if they move around so that the mask always stayed lock to that element. (Watch the Digital Intermediate featurette on the Fellowship of the Ring DVD to get a glimpse into what’s possible.)
Stop/Eject was graded by Michael Stirling at the company he runs in East London, White Cross Post Production. He very kindly put in a lot of hours, including two evenings, to make sure we got the best out of the film’s images. Even at this late stage in the game we were still telling the story: drawing the viewer’s eye to critical elements in the frame, enhancing lighting transitions when time stops and starts, making the happy past sequences warm and inviting, and contrasting those with cooler, darker scenes in the present.
Given that we shot in 16:9 (standard widescreen) but were cropping to 2:35:1 (cinemascope), we had the opportunity to move the images up or down behind the widescreen mask to reframe shots slightly; this is known as re-racking. We also added some subtle 35mm grain to the whole film.
The grade was the first time – and perhaps the last – that I was able to see the film on a really high quality, properly calibrated projector. It really exposed the quality of the camera and glass that were used. And while the Canon 600D and relatively cheap lenses look great on every other screen I’ve watched it on, on Whitecross’ projector I could really see the value of investing in high-end lenses. Even the difference between the Canon 50mm/f1.8 (£60) and the more expensive Sigma 20mm and 105mm lenses was apparent.
But I digress. The important things are: the grade looks great, and Stop/Eject is now finished. Hooray!
Sophie and I are off to Cannes this weekend – subscribe to my YouTube channel to get our daily video blogs. And when we get back it’s film festival submissions, DVD/Blu-ray authoring, and premiere arranging all the way.