Linking Sound Devices audio files with video in Final Cut Pro. . .

[Stephen Pruitt]

Hi all. . .

At one time, I was thinking of buying a SD 744T four-track recorder. Then I read that the files that the 744T creates aren't true .BWF files and therefore won't "auto-link" with timecoded video in Final Cut Studio 2.0.

Why would SD make a machine that doesn't create automatically linkable files???

Is there some EASY solution around this problem? It seems as if I'm going to have to record all of my audio in the camera only. I am NOT going to manually like up all those audio files shot by shot. I'll have 10,000 shots in this feature. . . a 1000 of which I'll end up keeping. That's a LOT of work.

:-(

Stephen

[Steve House]

Originally Posted by Stephen Pruitt

Hi all. . .

At one time, I was thinking of buying a SD 744T four-track recorder. Then I read that the files that the 744T creates aren't true .BWF files and therefore won't "auto-link" with timecoded video in Final Cut Studio 2.0.

Why would SD make a machine that doesn't create automatically linkable files???

Is there some EASY solution around this problem? It seems as if I'm going to have to record all of my audio in the camera only. I am NOT going to manually like up all those audio files shot by shot. I'll have 10,000 shots in this feature. . . a 1000 of which I'll end up keeping. That's a LOT of work.

:-(

Stephen

Don't know where you got the notion that SD recorders don't produce 'true' BWF files, your post is the first I've heard of it. I expect to be at an SD dealer later this morning and I'll check to see if they know anything about this. I'm wondering if you might be a victim of the confusion sometimes encountered between the BWF file FORMAT and the BWF file EXTENSION - the standard extension for a BWF file is WAV, same as a conventional wave file, and not BWF as used to be encountered a while back. I think originally the SD recorders used the BWF extension but the current firmware writes its files with the WAV extension - they're still BWFs though and have the TC generated timestamps. If FCP requires the old extension, just rename the file after copying it to your computer.

You didn't mention what camera you're using but the most common problem is that the timecode recorded with the video and the timecode-based timestamp recorded on the BWF audio file doesn't exactly match due to the inability of the camera used to send its code to, or receive external code from, the audio recorder. If the code produced by the TC generator in the camera and the code produced by the TC generator in the audio recorder aren't slaved to each other somehow so they are identical to the frame, an NLE obviously isn't going to be able to automatically sync the video and audio together. I'm not a Mac user so I don't know what might or might not be possible in Final Cut with regard to handling BWFs or how one goes about locking an audio track to a video track in it but AFAIK, the files produced by the SD recorders are perfectly normal, industry standard files. Now it's true that they now have MP3 and FLAC as optional file formats and if you've recorded as FLAC you need to use a converted to turn it into a BWF but that's another issue.

[Wayne Brissette]

To answer Steve's concern... You simply take the audio file and open it in FCP. FCP places all audio tracks on the timeline. There are some gotcha's and Steve touched on these. Time code is the biggie!

First, you must set the time code on the recorder to match that of the camera. So, if you're shooting true 24P, you want your time code to be 24 FPS. Normal video is done at 29.97 NDF for NTSC. For real film, use 30 NDF. This can be pulled up or pulled down if needed, but in general it's best to use the correct time code from the start.

When shooting, in addition to recording on the 744T, you want to send a mix to the camera. Last month I knew the final product was going to be put together in FCP, so I sent time code from the Deva to one channel I was feeding the camera. We used the Deva's time code as the master time code source, it was used to jam the slate, and since we shot on the Panasonic HVX2 no time was jammed there, which is why I sent time code out on one channel. One thing I have learned about FCP since then is it needs some additional software to be able to read the time code on the audio channel. So, basically what we shot didn't have time code, but the editor still had some easy ways to sync the audio and video. On both the 744T and Deva, time code is placed in the broadcast wav file and so once those are imported into FCP, they are placed on the timeline according to their time code. Since we jammed the time code slate, everything can be matched up using the information on the slate. The audio sent to the camera, is used as the guide track and once you align the video and audio, you can delete the guide track (or simply disable it in FCP). So, while perfect, this workflow is actually pretty simple.

On the topic of file names, .WAV, however as Steve points out .BWF is sometimes still used, and when I used the 722 on a shoot a couple of months back, the files came out as .BWF, so I don't know if this is a SD thing, or what, but they really should be .WAV. If you know the final product is going to be used in FCP, create your audio files as poly BWF files. This puts all the audio channel into a single file. FCP places each channel into a separate audio channel from the poly file. Some applications still don't like poly files, so you may have to use a utility afterwards to make each channel a separate file, but it is easier if you can, to simply leave the audio as a single file.

My final thoughts... Don't be an idiot, use a time code slate!!! If you don't, you're right you'll have to use all the audio sent to the camera because you won't know what audio goes where. There is no reason to use a 744T if you're not going to ever use the 24-bit audio it is recording and the 744T makes a horrible mixer, so you don't want to use it simply as a mixer. At a minimum, if you don't use a time code slate, use a dumb slate. Call out every scene and take, so you have a record on each audio file. Use the clapper to help you align your audio in FCP (along with your guide track). Have the sound person give you a daily sound report along with a CD/DVD of all the audio used for each day. The sound report should include every file, scene and take associated with it, and time code if possible.

You may think this is too much, but this is how we do it. I've done this on features, shorts, and industrial shoots. This is what works. I've had one director of a short who didn't slate ("it's not how I work")... I worked with him recently on an industrial shoot, guess what? We used a slate because he was never able to use the audio I gave him from the Deva on the short. Basically, he learned a hard lesson. When using a recorder, use a slate and save yourself some major headaches later.

Wayne

[Steve House]

Originally Posted by Wayne Brissette

...
On the topic of file names, .WAV, however as Steve points out .BWF is sometimes still used, and when I used the 722 on a shoot a couple of months back, the files came out as .BWF, so I don't know if this is a SD thing, or what, but they really should be .WAV. If you know the final product is going to be used in FCP, create your audio files as poly BWF files. This puts all the audio channel into a single file. FCP places each channel into a separate audio channel from the poly file. Some applications still don't like poly files, so you may have to use a utility afterwards to make each channel a separate file, but it is easier if you can, to simply leave the audio as a single file.

...

Just FYI - starting with firmware 2.09, Jan 12 2007 release, SD dropped the BWF file extension option. Since then, the files always have the BWF data chunk but only use the proper WAV extension. If the recorder's firmware hadn't been updated (now up to version 2.40) it would be still using the old system.