When the audio postproduction process of a movie starts, I am frequently asked by editors or animators in which formats they need to deliver the data. As there are often compatibility issues between the editing software and the audio DAW, here is an overview of how the data has to be ideally formatted to avoid those issues, if working in Pro Tools:
Timecode:
- the videofile should have a visible timecode in the frames, ideally in the top right corner highlighted with a black background - in this spot it shouldn't cover any sound-relevant information like faces or footsteps
- the movie starts at TC 10:00:00:00
- at TC 09:59:58:00 (which means 2 seconds before the movie starts) there has to be a 1 frame long peep synchronous with a 1 frame long test pattern
Video-format and codec:
- DV-Pal is a codec that works great in pro tools, as well as Motion-JPG-A - but it is important that all frames are exported as Keyframes, as otherwise the sound may be asynchronous in the end
- A Codec that will definitely NOT WORK is h.264 - Pro Tools is not able to process this codec, as the compression is too heavy, which results in a bigger workload for the system to reproduce the frames
- Resolution: 720p should be big enough to see what is needed and the file size is limited
- Framerates: Please transfer the video in the framerate that is also needed in the end. That means (in Europe) 24fps or 25 fps or both, if both is needed. Different framerates result in different video-lenghts and the audiofiles have to fit, what means being stretched/shrinked and tested with the right videofile.
- Format: Quick-Time Movie (.mov)
- Progressive is always better than interlaced
Audiodata:
- The audio tracks should be exported as an OMF (2.0) session
- There are two possibilities when exporting OMFs: export embedded files or referenced => "Embedded" formats have one big file containing only all the USED audio data, that means the folder will be smaller, but you have one big "data" file to work with. "Referenced" means, all files that have been used in the session are copied in their original status and length, which means the folder will be bigger but you will really receive everything and can search for alternatives in the files, if a take in the session does not work as desired.
- It is up to your editor and you which way you want to go, but always keep in mind, that if you work with embedded OMFs, you need a "Handle-Lenght" that is as long as possible. The handle length can be set in the settings of the editing software and should at least be 500 frames or 20 seconds long
- And dear editors: please do not delete any audio tracks that have been delivered by the production sound mixer, no matter how awful they sound - always just MUTE them. When files are muted your post production supervisors will also receive them muted and therefore will know what you wanted, but he can still access the file, if needed
I hope this will help some of you, if you're not sure how to deliver the session.
Best regards,
Marie
