I have pro res 422 footage shot on a 7d at 25fps, I have wav files recorded on an hn4 recorder. The audio sample rate of both is 48khz, the waveforms appear to match and yet in FCP 7 the audio begins to drift out of sync after about 30 seconds.
I've tried re-opening the project and changing the set up, I've tried re-sampling the audio, I am now out of ideas!
Timing audio with DSLR footage is a pain sometimes. There are apps - most notably PluralEyes - that can handle retiming this for you and it's really easy to use. I don't own it, but I have had a chance to use it in a real-world situation. http://www.singularsoftware.com/pluraleyes.html
I generally do my sync manually with short clips and short audio and I don't usually have issues - make sure to keep really good notes during production to shorten the process of locating clips and audio. With longer clips, you need a visual in and out as well as an audio in/out. Set markers in FCP for the visual and audio in/out points, get the timing between each one, then adjust audio scale to match video. I haven't done this in PAL or 25fps, but it works for me with 24P and 30P footage. It's a bit of a pain if you have a lot of clips, but it should still work.
My in/out points are usually when the clapper starts the scene and when I yell "cut" - for visual cues, it's the clapper and usually a bit of dialogue or mouth sound just before or after cut because the drift will usually be apparent at the end of the clip. I should start doing an end clap on camera...hmmm.
In the future, record at 96kHz and 24-bit to ensure the highest quality. That's what I do with the H4, but I haven't upgraded to the H4n yet so the settings may be a little different.
[Jonathan Ziegler]"In the future, record at 96kHz and 24-bit to ensure the highest quality."
Digital video is 48khz so there is nothing to be gained by recording at 96khz and then having to convert everything back to 48khz before editing. Just adds an extra step and you end up back at 48khz regardless. It also uses much more storage space in the field. 24 bit makes sense to get extra headroom.
Most of the audio drift posts here at the Cow are with NTSC frame rates not 25. Was your Easy Setup on a 25 fps setting when the wav files were imported in? Have the files been imported via itunes or third party software or imported directly as files into FCP. Did you import or just drag and drop the files into the timeline?
Thanks for this Jonathan, I'll give PluralEyes a look.
Interestingly, the problem was solved by altering the speed of the .wav files to 99.9%.. I can't quite get my head around that, but I am past caring too. They start in sync and over the 10 or so minutes of each one, no longer drift out at all.
Sweet! I hope it works for you. I used to always think it was the dropped frames in NTSC, but since I've heard this happening with PAL as well as NTSC, it must just be a very slight difference in timing from the video and audio - losing .1% every 10 min? A scant few seconds? Since neither device will do a time code, you can't really sync separate files and must fix it in post.