Red cam music video shoot: creating playback mov for video iPod

[Jean-christophe Boulay]

“How do I convert the audio to 23.98?”

You don’t. You’re talking either video frame rate or time code rate. Audio has sample rate, in your case 44.1kHz. The rest of the workflow seems pretty iffy to me, but if that’s the way it has to go down, I guess it could work.

Ideally, you’d really want your audio on a timecoded source, to be able to run that TC to your camera. Some major swearing from the video editor would be saved this way. Traditionally, music videos are shot using an audio CD with the song in mono on the left side (that gets played on the set) and audio timecode on the right track (that goes into the camera TC input). The TC format you use should be the one you shoot in. That way, every time the song is started, TC goes to the camera and all takes can be aligned to audio for editing in seconds, without having to go back to the beginning of the take all the time. I’ve been involved in such projects countless times and this workflow has always been airtight.

IHTH

JC Boulay
Technical Director
Audio Z
Montreal, Canada
http://www.audioz.com

[Thaxter Clavemarlton]

[Jean-Christophe Boulay] “every time the song is started, TC goes to the camera and all takes can be aligned to audio for editing in seconds, without having to go back to the beginning of the take all the time.”

Couple of ways to make that work better for you.

1- Change TAPES with every take. Each tape can be captured separately.
If you need to recapture the tapes, it’s a piece of cake.

-or-

2- If you do record multiple takes on the same tape,
be sure to carefully SLATE each take and then
capture the tape manually, capturing each take as its own separate clip.

And, if you have, say, 10 takes on the same tape, each with the same time code repeated over-and-over, and need to RE-CAPTURE the tape (to Up-Rez, or change editing locations, or for any other reason) and simply stick your one (multi-take) tape into the machine to batch capture…
Oh, man! You’re in for a mess! The edit system can’t tell one take from another…
They all have the same TC.

And, allow plenty of extra TC at the head of the CD before the music track begins.
CD should have some sort of audio countdown or click-track that counts the musicians
into the head of the song.
(I like about 15-30 secs. of good solid TC before the song starts.)

[Steve Watkins]

interesting stuff guys.

Well we’re not recording to tape. The red records to a hard drive.

We’re doing this on a D.I.Y. budget so there’s no way we’re going to be able to get a time code generator and send TC to the camera from playback. Or get a jam sync slate.

I have to work with what’s possible for us right now.

I know that there is time code for video and sample rates for audio. what I’m talking about is the shift that occurs having the band sing to audio recorded at 24fps drop frame. the footage is dropping a frame which effects the speed of the band singing and playing so by the end of the song there will be significant drift from the original audio track.

Therefore I’m lead to understand that i need to speed up or slow down the audio by 0.1 % or something like that.

anybody else out there do playback for music videos?

[Ty Ford]

Hello Steve,

Why 26 FPS drop frame. Why not 24 non-drop?

Regards,

Ty Ford

Want better production audio?: Ty Ford’s Audio Bootcamp Field Guide
Watch Ty play guitar

[Steve Watkins]

we are shooting in 24 fps

[Richard Crowley]

“That should work right?
How do I convert the audio to 23.98? ”

So the iPod playback and the sync audio track on the Red are only for reference, right?

The REAL music track from the master mix is layed into the NLE. I’ve done this many times with no timecode, just by manually tweaking the video into place so that the video (scratch) audio track matches the (master) reference audio track from the original music mix. It isn’t rocket surgery. Takes longer to explain than to just do it.

[Steve Watkins]

maybe it’s just me but it seams pretty condescending around here. Manners. They don’t require a rocket scientist either.

[Thaxter Clavemarlton]

Drop frame and Non-Drop Frame are the exact same “speed.”

With DF, no actual recorded FRAMES are dropped.
Only certain of the Time Code frame NUMBERS are dropped.
But, even in DF, every single video and audio frame remains on the recording.

DF and NDF are simply two different ways of NUMBERING the TC track.

DF will yield a TC that is clock-accurate, NDF will yield a TC with no numbers “missing.”
But neither of those will in any way affect the speed, quality, or length of the recording playback.

[Will Salley]

I do music video a bunch.

Here’s how we do it most of the time:
Single camera (or multi) w/ timecode in from playback device ( Sound Devices 744T) with final mix on the drive.
Music track is 24/96 from Pro Tools with a 30 second countdown and click.
Timecode is 29.97 NDF throughout the entire project. I consider 24 FPS too stroby for the fast movement of most music projects. We’ll usually shoot 30P and even overcrank to 48 on non-synced shots – especially if the camera is on a boom – which it usually is.
(even with super 16 film, we shoot at 30 FPS and open the shutter angle to give it some blur)

The playback audio is split out to the cameras- just in case.

The same audio track (WAV with embedded TC) is brought into the NLE and maintained.

If you don’t have access to a TC capable deck or playback, do it this way:

Create a project in the NLE, import the master audio track and place on the timeline. LOCK the audio track so it can’t be moved. Use a timecode generator filter on the track and make the TC window as large as you can fit in the preview window – well give it a little room on the edges. Then render a 1280×720 (with audio) Quicktime of that. Use that as your playback file in a laptop. Shoot the laptops monitor at the beginning and end of each take. Sync visually in post using the same project that you originally created. You should still be able to use Multicam in FCP or Avid to cut with, which is essential for cutting music videos.

Mac Pro 2×2.8 Quadcore – 10.6.2 – QT 7.6.3 – 22 GB RAM – nvidia8800GT – SATA internal & external storage – Blackmagic Multibridge Pro – Open GL 1.5.10 – Wacom Intous2 tablet – AJA io
SONY XDCAM EX3 – Letus Elite

[Richard Crowley]

Sorry, I wasn’t implying anything about you personally. It just seems like people assume that sophisticated timecode and special equipment is required for syncing music-video-type production. I’m no genius, but it just seems pretty simple to slide the video clips into place against the master audio track. Music videos are a special case that make this a lot simpler than more conventional types of production.

If you have the equipment available, do the experiment and see how it works. I just think it is simpler than many people assume.