Is there a way to remedy this without ungrouping the video from the audio...

Not that I've found.
I've had to do this a few times when I videotape a play and get a feed from the audio board in addition to a second track fed from a shotgun at the back of the hall (ambience & applause).
I turn off "Quantize to frames", set up a small loop (10 sec.or so), set it to repeat, zoom in on the timeline and then slip the audio track by using the 4 & 6 keys on the numeric keypad until the sounds match.

Mike

The speed of sound is 1000 ft per second. Divide that in frames and it's roughly 30 feet per frame when shooting 29.97. So you are talking about being one frame off.

You should select a few seconds of your multicam and do a ram render to check for sync. It shouldn't take that long to do.

Also I just thought of something that might work on location. Normally we try to sync cameras by using one loud sound, such as the bang of a film clapper. For multiple cameras where one is very far away what if you used two loud sounds. Or even better, a bright Flash followed by a loud sound. The bright flash should reach all cameras instantly. You could then measure the number of frames between the bright light and the loud sound.

You could set off a still camera flash and follow that a second or two later by a clapper.

The speed of sound is affected by temperature. At 72f, it travels at 1130.809 feet per second. This works out to 37.731 feet per frame at 29.97fps. For other temperatures, you can calculate it here:

http://www.sengpielaudio.com/calculator-speedsound.htm

1. Ungroup the audio from the video.

2. Slip the audio to match other tracks (press the Alt key to "slip").

3. Re-group with the original video.

The audio will be pink, since it is no longer in sync, but it will stay in the same relation to the video as you make further edits.

I have a clapper with a small camera flash wired to it for use in multi-camera concert shoots (cameras near stage and at rear of hall), when you snap the clapper the strobe fires. I sync the video using the strobe flash, and then sync the board audio by matching the audio spike to the flash on the near-stage camera. If I was mixing in hall sound from one of the cameras for ambience I'd match that audio spike up too.

I may be missing something here but I use a master audio source for all the audio and then this is not an issue. You are cutting between the audio as well from one camera to the next? Makes for uneven audio sound quality. Do you have one camera that can provide the audio for all the sources?

Gary

Gary, I can only speak for myself but I use two audio tracks a lot when I videotape plays.
I do that because I generally get program audio from a house PA.
Unfortunately, most board ops kill all the mics at the end of a scene which means that I don't get any applause from their feed.
This is where the shotgun comes in handy.
The other thing I notice is that the ffed from them tends to be somewhat dry so adding just a bit of room ambience from the shotgun mic helps to give it that "live sound".

Mike

I may be missing something here but I use a master audio source for all the audio and then this is not an issue. You are cutting between the audio as well from one camera to the next? Makes for uneven audio sound quality. Do you have one camera that can provide the audio for all the sources?Just to add to the last post, I too use the sound board (when I can get it). However, there are two more reasons to want to get good sync on all audio channels. First, I often mix about 10% of one of the live mics into the front channel, and 100% of that same audio into the rear channels for the 5.1 mix. It provides a much more realistic experience. The soundboard audio is often too "studio" to sound like a live performance.

The other reason is lip sync. When I do dance, this isn't a problem, but when people are speaking and you cut between cameras, you'd better have the audio synced. If you use a flash or visual element to sync the different cameras, this obviously won't be a problem, but if you try to sync using the audio from each track without first slipping the audio from the distant cameras, you'll have problems. And, if the soundboard cuts out -- either because they fade or because of a problem -- you can simply fade to the audio from the other track without having to remember to do any further work.

As I mentioned a few weeks ago in a similar thread, I get my focus points during rehearsal, and note those distances. I then use these focus presets during filming, but also use them (along with the math presented by others in this thread) to slip the audio the "correct" amount. Then, as was pointed out by several people in that other thread, I'll tweak that slippage a little using my ears in order to make it sound right (echos sometimes require a little fudging).

I'm trying to grasp the instructions above (1. Ungroup the audio from the video. 2. Slip the audio to match other tracks (press the Alt key to "slip"). 3. Re-group with the original video.), but I must be getting old. I understand how to FOLLOW each step individually but am confused as to what I would have done before Step 1.

I synced all 4 tracks last night using the audio, so the audio already matches... both visually and when listening. But obviously the video from the rear cameras is slightly off. Let me see if this helps; here's the setup I had:


A Cappella Groups (10-12 singers)

Cam 1 (around 12-15' away in center)
D50 recorder


Cam 2 (30' back, 30' to side) Cam 3 (same)



Cam 4 (approx 50' back, 25' to side)

Cam 1 and the D50 are very close, and I synced easily with a little Ctrl + drag to the D50 audio event. I've already synced up the audio in Cams 2-4 to Cam 1 & D50. But as stated, the video is a few frames off. If I were to start with Step 2 above, would I then slip the video somewhat rather than the audio?

over the weekend i saw a pbs concert with the steve miller band,,, awesome cept they ran into the same issue you are having only didn't fix it,, audio was out of synch on at least half the shots,,
Havent read all the posts see a lot a math and it gives me a headache,,, i would simply lay the audio track , lay all your cams and do your visual edit,then tweak the shots to the master audio

Since light travels at 186,000 mph, distance is not a factor visually. You want to sync all your cameras visually first. When you sync the video tracks make sure the "Quantize to Frames" is ON. Moving the video by a fraction of a frame will cause all kinds of problems.

At this point you will want to sync your D50 master audio track to the video. The D50 will probably have been recording at a slightly different rate than the video, so it may require that you time stretch the D50 audio track to make it fit the video. Then unlock the audio tracks from the various camera tracks and sync their audio to the D50 audio. When syncing the audio, you can turn quantize off.

Since none of your four cameras nor the D50 were synced to a master clock, it is likely that they will drift apart from each other over time. Crystal controlled devices like digital video cameras and audio recorders are very stable and their individual clock frequency won't vary over time, but no two devices are operating at exactly the same to begin with, so they will drift. How much they're going to drift depends on how close their clocks were to the same frequency and the length of your program. You may get lucky and they will be close enough to pull it off without much trouble.

Download this file and drop it on your browser:

Tutorial

It is a tutorial, in Flash format, created with Wink. I tried to add audio, but Wink is pretty primitive and I couldn't figure out an easy way to dub audio after the fact.

Anyway, you'll see the two cameras, and note that the audio doesn't match, even though the video has been perfectly synced using a camera flash that happened during the performance. I first ungroup the audio on the first camera, then press and hold the ALT key and then "slip" the audio to the left. I then re-group the audio to the video. At this point, the video of the two cameras is exactly the same as it was before (i.e., perfectly synced), but the audio now matches.

If you really want "perfect" sync, then set up a 2-3 second loop, and while playing back the looped audio, turn off "Quantize to Frame," ungroup the audio again, zoom way into the timeline, and then while holding the ALT key, slip the audio back and forth a few milliseconds until you get phasing and other audio artifacts the way you want them. Don't forget to re-group the audio and video before you continue editing.

Aaaaaah..... from John Cline and Johnmeyer, I think I get it. Sync the VIDEO first. THAT is what I was missing. Sorry for being so slow. I actually get the group/ungroup steps and the slipping technique and have used those a bit, I just wasn't getting it that I needed to sync the video and then do the steps indicated. I'm going to work on this a bit tonight.

I've actually done a number of concerts over the past few years, but they end up like the Steve Miller footage mentioned above, and with each project, I'm trying to improve the quality and pick up issues like footage a frame or two off. Thanks for all the excellent replies above.