How to Fake a Crane Shot for Nothing
The edit ran long, as edits do. The first assembly came in at nearly double the final length and included an entire sequence that we still think is good and still had to remove.
Gallery 12 photos · 2 MB
By the third day we had stopped fighting the weather and started building the schedule around it, which is the single most useful thing that happened on the whole production.
Notes from the edit
Sound was the part we underestimated. It always is. The camera work forgives a great deal more than the mix does, and audiences notice the difference long before they can name it.
- One key, one bounce, and whatever is already switched on in the room.
- Meter for the highlight you cannot recover, then let the shadows fall where the scene wants them.
- Match colour temperature at the source rather than in the grade.
- Leave one practical slightly hot so the frame has somewhere to breathe.
You do not get a better film by adding equipment. You get a better film by removing the reasons a shot is not working.
How the shoot was structured
None of this is prescriptive. It is what worked in one room, on one shoot, with one set of limitations. Take the parts that map onto your own and ignore the rest with a clear conscience. The full rundown lives on the about page, including the gear we no longer recommend.
- Block the scene with stand-ins while the lights are still cold.
- Set the key, then kill every other source and add back only what is missing.
- Run one take at full speed before adjusting anything else.
- Shoot the wide last, once the performance has settled.
The transcode command, for anyone who wants it verbatim: ffmpeg -i master.mov -c:v dnxhd -b:v 36M proxy.mov. It is not clever, it is just consistent.
card_01/ ingest verify proxy backup card_02/ ingest verify proxy backup card_03/ ingest verify -- --
One last thing. What follows is the part that is normally cut out of these write-ups: the setups that did not work, the compromises that made the schedule, and the two decisions that ended up carrying the whole piece.
There is a version of this where we hired a bigger crew and solved most of these problems with money. It would have been faster. It is not obvious that it would have been better, and it certainly would not have been finishable.
Comments (0)