reconforming audio on a multiclip or multiclip sequence

[Andrew Gurney]

Create your multi-clip as you did orignally. Assuming you had 3 video angles and 1 Audio Track you will have a multiclip with 4 windows.

In the viewer, click on the middle button (Next to the Scaling Properties) and make sure Video + Audio is checked.

Back in your multiclip, click on the window that holds your audio track. The Green and Blue Outline should now be around your audio clip.

Now click on the middle button again (Next to the Scaling Properties) and Check just Video.

Essentially this has locked your multiclip to your audio track so that whenever you switch video angles your audio will remain on your production audio.

Also because you have video clips and only one audio clip there are a couple of ways to do it. First way is to create a multiclip with all of your video angles and use as is. It works but because your multiclip has so many angles it can often be unmanageable. The second option is to create several multiclips all of which have the same audio track but where each multiclip has a different set of video clips. It may seem redundent to have so have some many copies of your audio but I find that it is a easier workflow

If you wanted to merge your audio you would need to do it before you create the multiclip but you would still be required to lock your audio as outlined above.

Hope this helps, it’s also explained pretty well in Final Cut Manual

andrew

[Andreas Kiel]

Jean-Christophe,

The way Andrew described wouldn’t work for you.
To make it work you need polyphonic BWAVs. You can convert your mono files to polyphonic using the free WaveAgent from Sound Devices.
But also that won’t solve the problem in way that you can rely on your multiclip editing.

Finally there is one save way to work with multiclips and external sound.

Before explaining it here some background on audio you have to understand.
First read pages 325 to 327 of Final Cut Pro manual – if not already done.

This explains a bit more of the BWAV handling.
If you are editing with NTSC the manual forgot to explain what happens, also the DF/NDF issue is really explained in a way that the user understands clearly.
So working with NTSC means you do use “slow seconds” – one second of video needs 1.001 seconds to play back. Audio in FCP never uses “slow seconds” that’s why several users got the experience that audio is going out of sync with longer clips.
The explanation of NDF timecode in the manual is a bit misleading as well. As described audio uses 30000/1001 as the 29.97 timecode, but audio can have 30000/1001 DF and NDF timecode which Final Cut doesn’t have. Also Final Cut uses since the early days 29.97 DF timecode and may handle the 30000/1001 frame rate but there are issues which are documented in the technotes on the developer site. So with BWAV time(code) calculation the accurate TC rate is used, with Video the rounded TC rate is used.
The most important thing though is that the audio does not have a kind of TC track which can be used within Final Cut Pro to keep video and audio in sync under any circumstances. As described in the manual, audio (if BWAV or Sounddesinger II format) does have a “time stamp”. This time stamp will be used to map a SMPTE timecode on the fly to the audio (based upon your settings of the FCP project). The displayed SMPTE TC is not a QuickTime TC track, but something like a “TC Generator Effect” you know from the video effects menu.
Back to NTSC. The manual only describes how a 48 kHz file will be mapped to a 25 FPS video and therefore it looks easy and straight forward. It does not describe what will happen with NTSC (not 29.97 DF nor any other NDF NTSC rates). They talk about a pattern which will be used – unfortunately it is not explained what the pattern is. There are multiple options for these patterns and they are used with MXF and other formats and the vendors do use the one or the other – bad NTSC world.
My experience, and this includes the experience of really many other users: FCP doesn’t use any of this patterns. So as already mentioned earlier NTSC Video or interleaved Audio/Video needs 1.001 seconds to play back 1 second of real world video. A 48000 Hz audio needs 1 second both in real world and in Final Cut Pro. The user Prefs for NTSC audio timecode (DF/NDF) are also not a big help here.
Here an example what will happen with 24 FPS means NTSC 23.976 FPS editing. After 24 frames 48048 samples of the audio are played – means the stuff is out of sync compared to the original 48 kHz recording. In real world people are often not aware about this issue because clips are mostly short and the out of sync stuff is not obvious. With a 24 FPS non NTSC editing every thing is fine.

What you can do is either to record not at 48 kHz or change the playback sample rate in the the file. But this can lead to problems if editing timebase is different from target timebase.
So another option is to transcode the audio file into a QuickTime wrapper with a QuickTime timecode track. This timecode track will hold the playback speed and therefore the “virtual sample rate”. So with the above 24 FPS audio this audio will behave correctly in any 24 FPS project regardless of being NTSC or not.

With this TC track we are back to your problem: creating multiclips the safe way.
One way and that’s the way we use (many others as well) is to sync the clips with external audio and create new files. If you want to link them in the timeline you have to unlink audio and video and than link the clips with this new combination of video, cam audio and external audio.
If you’re done, export all of these new clips to Quicktime movies (assuming that video TC and audio TC do match). Unfortunately you can’t do a simple offline, online with the video clips you already have in the project – FCP will give you an error message about the non matching amount of audio tracks, but you can do it with the linked/merged clips
Another option is to create a new Browser Column Layout which doesn’t include the audio/video track layout but all metadata columns you need. This can be exported as Batch List, then be imported into the project and used to re-connect the files.

From there you can go and create multiclips which will stay in sync in any case and behave like multiclips created from standard camera clips with common timecode.
If you want to follow Andrews advice you can link the audio tracks in the timeline and do the same steps as above with the linked audio clips.

The drawback of this method though is that if you want to go back to Pyramix (or any other DAW) using the original sound files it won’t work, as they are “caught” in the QT wrapper now. If it’s “only” Protools you can use OMF.

We do this sync and multiclip stuff on a daily base and I think I have seen any problem which can happen and therefore I have built a custom application which does handle all this stuff outside of Final Cut Pro. This app does handle the creation of “merged” Quicktime movies different. These movies allow to retrieve the original sound and can be used in any project. I hope I can make it available to the public in a few weeks – but it will be quite expensive (you can send me an email to kiel(at)spherico.com if you would like to be informed when the app goes to a private testing status).

One thing which is currently free is my SequenceLiner which will help you to create a multitrack timelines where all clips are in sync (as long a everything is recorded with free-run or TOD TC). You can use this as base for the steps I described.
If you happen to attend the LAFCPUG Supermeet during NAB forum moderator Jeremy Garchow will show how to setup a five camera (P2), 8 track external audio recording session using SequenceLiner and mxf4qt.

Hope this helps to make things a bit clearer.

Andreas

Spherico
https://www.spherico.com/filmtools

[Lori Loop]

Hi there, I love use multiclip for my projects with final cut but I can’t understand how I can deselect the green & blue outlines (means video & audio selected in my multiclip). Can you explain me how I can render a project that doesn’t show any outline? thanxx! lori 🙂