Bringing BWF files into Final Cut Pro 4.5
I shot a short film last week with very limited resources, but we decided to rent a PD-6 and get good sound on separate channels, well-mixed.
I have downloaded the files from the PD-6 to my computer and now have, per take, one 4-track file for lavaliers and one 2-track file for the boom mike.
Two programs have been recommended to me, in order to bring these files into Final Cut - Sebsky Tools and BWF2XML. With both of them I have the same problem:
The 2-track file imports into Final Cut Pro just fine, and can be brought into the timeline with no problems. However, I can not view those clips in Viewer or Canvas in order to fine-tune level settings or find peaks, etc. They open, but the sound tab doesn't show any waveforms.
The 4-track file imports into Final Cut Pro and are visible in the browser. I can not view them in Viewer or Canvas, as above, but I also can not render them in the Timeline - they lay in as 4 tracks, but will not play.
I was told that Fostex had a Waveform Manager (WAVManager Ver2.00) program that would parse out the tracks (into single files? I'm not sure....) to assist with that problem, but I downloaded the program and it seems to want to import from an MR Drive (?).
I was then pointed to a program called Audacity by a friend - I can bring in the 4-track file, and in turn delete all but one track, save that track as a unique file, until each track in the original file has its own file.
This system works fine. The files import into Final Cut Pro, play in the timeline with no rendering and the waveforms are visible in the Viewer and Canvas.
But this process is very cumbersome, even though we have a very short project. I can not invest in extra hardware and don't own ProTools, so I need a freeware solution, which I was given the impression there was. Is this the best I can do?
I have the additional problem, by the way (I am a camera person and am very ignorant about sound, although I have learned a ton in the past week!) that we accidentally had the PD-6 set at 24 fps. Initially, I was told that we'd have to transfer all the files from the original PD-6 hard drive we used to another PD-6 that was set at 29.97 (or 30? I'm editing in NDF). But I'm finding with takes that are up to one minute that the sync is fine by eye, and with longer takes I figure I can resync at minute intervals to be safe. Any thoughts on this part? All three programs mentioned above, btw (Sebsky, BWF2XML and Audacity) seem to be solving part of that problem by restriping to NDF timecode, and so I'm gathering that then I'm just fighting the smaller pull-up problem. I'd appreciate any input on this.
Many, many thanks from a newcomer to this particular forum.
I should add that we are going to export to DigiBeta only, never have to worry about a 24-frame output.
I think your best solution is to upgrade FCP to version 5.1.2, which will read the .bwf's natively (and display the waveform). It eliminates that extra step that is required with those other tools you mentioned ( I used both Sebsky and BWF2FCP until 5.1.2).
Did you shoot on film or video? These days the term can be used for both mediums. If video, did you sync the PD-6 to the camera or vice-versa? At what frame rate?
The BWF timecode would be useful only if the takes were slated with a timecode slate OR sent to a video camera with timecode synced. Audio timecode recorded at 24 fps is used for film shot at 24 with the intent of printing back to film for theatrical release. Most every other useage calls for 29.97 (which can be DF or NDF) and is usually determined by the picture editor.
If all else fails, or is cost-prohibitive, you can always use the slate clapper.
System Info - G5/Dual 2 - 10.4.8 - QT v7.1.3 - 8GB ram - Radeon 9800Pro - External SATA Raid - Decklink Extreme - Wacom 6x8
Thanks Will - I appreciate the reply and the clarification. Just to answer some of the questions:
We shot with my Panasonic DVX100B and it was only later that I understood the capturing at 24fps but taping at 29.97 or 30 dichotomy. So, I wrongly set the PD-6 to 24 fps.
We didn't sync the PD-6 to the camera or use a timecode slate. We clapper slated every shot, hence the decreased cause for concern.
I have since settled on the solution where I'm taking the two-track files in through BWF2XML and I'm processing the three-track files in Audacity to send each track out as a unique file which I can then import easily into Finial Cut. I am still having trouble seeing waveforms in the BWF2XML imports. So, any advice on that would be greatly appreciated. But with both Audacity and BWF2XML I am converting the timecode to 29.97 and there seem to be no sync issues.
I am now considering upgrading to FCP5.1.2. and this will definitely accelerate that decision.
just found this thread by chance and thought it would be useful to give some comments (as I'm the author of BWF2XML).
If you don't need metadata support and you are not working within a NTSC project just import the files as they are.
If you are working with NTSC there might be synch problems with any kind of direct BWAV import since any version of FCP only supports non NTSC timing (NDF only) during import.
Depending on your audio setup and the later needs you may use the FCP 5.1.2 generic import option.
Here in short a explanation how it works with 5.1.2 (and QT 7, which needs to be installed, there also a doc on the Apple site, which is unfortunately not completely correct: http://docs.info.apple.com/article.html?artnum=304660:)
FCP will take the "samples since midnight" found in the BWF, the sampling rate of the BWF and the audio/video setup found at launch time. These three values are used to create a SMPTE timecode - which may be the correct one. It will never be correct with audio recorded with DF timecode, since FCP currently only supports audio NDF timecode during import.
FCP 5.1.2 allows to switch to DF after import, but you have to know the original timecode and modify the tc string to match the original timecode of the file.
As none of the versions of FCP does read any metadata (other than "samples since midnight" with 5.1.2) and due to the way FCP reads files, any reel information (or other information) has to be edited manually after import.
The timecode value is only held in memory (project) - so if you have to take the files offline, you may get a different timecode when re-connecting. Though this only happens if your launch setup of FCP is different from the launch setup of FCP when you imported the BWAV files.
Finally Sebsky or BWF2XML is the better choice in some cases.
To learn more on FCP 5.1.2 BWF import features have a look at http://www.spherico.com/filmtools/BWF2XML/BWF512.html
To have a nice test application to check BWF file settings against FCP launch settings download: http://www.spherico.com/filmtools/BWF2XML/BWFReader.zip
Tel.: +49 (0)721 183 9753
eFax: +1 650 897 8094
http://spherico.com/filmtools -- some workflow tools for FCP