Recording sound separate and filming multi-cam angles....

[Doug Young]

Originally Posted by KevWind

Yes I realize that Final Cut can sync the video to the audio. I have done it several times using multi cam shots in a studio type set up while simultaneously recording into Pro Tools. Then importing the mixed PT file back into FCPX and syncing it to the camera audio. But as you pointed out, how well that will actually work when recording four different takes with the attendant four different timing anomalies in both the video and the audio is the question

But if he records the original "real" audio to each video take, they should all be perfectly in sync. If he records multiple separate takes, with totally different audio, they won't sync, he'll just have to slice and dice and assemble, same as composing multiple audio recording takes.

[KevWind]

Originally Posted by Doug Young

But if he records the original "real" audio to each video take, they should all be perfectly in sync. If he records multiple separate takes, with totally different audio, they won't sync, he'll just have to slice and dice and assemble, same as composing multiple audio recording takes.

Yes I agree I don't think I said other wise. And I agree that getting the GB audio into the different video takes would be preferable.
I was offering a way to possibly make the four different takes (the way he described and if he was going to continue to do it that way) a bit easier by counting in and using that count in wave form peak to manually time slip each entire video clip to that start point. And then see how the 3 videos looks in comparison to the original GB audio

[Doug Young]

Originally Posted by KevWind

Yes I agree I don't think I said other wise. And I agree that getting the GB audio into the different video takes would be preferable.
I was offering a way to possibly make the four different takes (the way he described and if he was going to continue to do it that way) a bit easier by counting in and using that count in wave form peak to manually time slip each entire video clip to that start point. And then see how the 3 videos looks in comparison to the original GB audio

Ah. Might work. My son's band recently did a video pretty much like this. They shot like 6 takes, not even playing along with any audio, then just synced all the shots with a well-recorded audio, doing the sync manually. Apparently their timing and tempo was good enough it all just works. I can't see anything out of sync at all. So it can be done.

Another approach would be to record audio+video, mutiple takes, but each one being both audio and video. Then use GB or FC to sync the audio video, and you'd have N different takes, each with good audio (different performances), and then just put them together.

Personally, I'd just go buy a few more cameras. Unless this is for some commerical release, $100 camcorders, or even a phone are plenty good enough. Then you do avoid all these hassles and just do a real multi-cam shoot.

[lawler]

So just to confirm -- would it be possible to shoot 4 different angles and insert the GB recording into each of them and in theory then use FCPX to make a true multi-cam which will sync to this implemented audio?

That would be an ideal scenario if possible.....

Thankyou again for all the suggestions - they are very much appreciated!

[Doug Young]

Originally Posted by lawler

So just to confirm -- would it be possible to shoot 4 different angles and insert the GB recording into each of them and in theory then use FCPX to make a true multi-cam which will sync to this implemented audio?

That would be an ideal scenario if possible.....

Thankyou again for all the suggestions - they are very much appreciated!

Yes, thats the idea. FC will sync to the audio, do if its there, you should be all set.

[lawler]

Originally Posted by Doug Young

Yes, thats the idea. FC will sync to the audio, do if its there, you should be all set.

So just to confirm (apologies for this incessant clarification!) the workflow would be:

Input one separate video angle into FCPX and my externally recorded audio.

Then somehow merge these two files together.

Save this file as an integrated video angle with audio.

Repeat x(n) based on how many angles I want to use.

Then grab x(n) videos and create a multi-cam video, which in theory I could then sync to the audio and go through a playback procedure where I can choose the viewable angles in real-time for the outputted video.

This would work perfectly if achievable - I suppose I would just have to physically try go through the workflow as no-one seems to have approached it in this way.

[Doug Young]

Well, that's not the flow I was suggesting. I suggest:

Record your "studio" audio - I assume you've already done that.

Now record one camera angle while you listen to the playback, playing along, while recording that same playback to the camera. You're listening in headphones, right? So you need a splitter that can send the headphone output to the camera's line-in. That means the camera's audio will be the original studio audio. Alternatives would be to play back the sound not over headphones, but over a speaker, loud enough that that's what the camera picks up, and you can hear it as well.

Do that as many times as you want with different camera positions.

Throw into FC and create a multi-cam clip. Select the studio audio as the audio for the mult-cam clip, and then switch at will between the different video options.

For sure, dive in and try things to nail down the details. You won't break anything. Do a 10 second clip first so you aren't wasting time before you get your workflow down.

That should result in a video where we get to see you pretending to play while we listen to your recording. I'm not normally a purist about such things, but I'm still stumped as to the value of that. If you don't want to record your performance "live", I'd be tempted to go all the way to a "MTV" style video where you create a story with the video that adds to the music. Doesn't have to be you playing at all, tho that could be a part of it. There's an interesting book called "From Still to Motion" that goes thru the entire process of creating a music video (singer/songwriter) with final cut. They even give you all the raw footage and audio and walk you thru assembling it into a video. It's intended for photographers moving from shooting stills to motion, but I think it's applicable to what you want to do, specifically because they demonstrate multi-scene shoots of a music video.

[KevWind]

Originally Posted by lawler

So just to confirm (apologies for this incessant clarification!) the workflow would be:

Input one separate video angle into FCPX and my externally recorded audio.

Then somehow merge these two files together.

Save this file as an integrated video angle with audio.

Repeat x(n) based on how many angles I want to use.

Then grab x(n) videos and create a multi-cam video, which in theory I could then sync to the audio and go through a playback procedure where I can choose the viewable angles in real-time for the outputted video.

This would work perfectly if achievable - I suppose I would just have to physically try go through the workflow as no-one seems to have approached it in this way.

Just to add to what Doug is saying the whole point of the "automatic" Sync to Audio feature in FCPX is that you sync all the video angles (or in your case takes with different angles) to the single GB audio (that is now in each take) at the same time.

This means the only variable is visually how closely you actually play each successive take to the same GB audio you are listening to each time . Because by inputing the same GB audio into the camera for each successive take you have eliminated the variable of how close the audio is between takes (which means FCPX can easily and accurately sync the different video angles/takes to the same single same piece of audio )
So you would in fact have all the camera angles and the GB audio in FCPX at the same time then perform the sync to audio the same way as shown in YouTube link that Doug originally posted.

[Doug Young]

Also, another approach is to just not use multi-cam, and not worry about audio on the video. Just line the videos up on separate tracks and slip them until they are visually in sync. Then cut the videos up and select which ones you use at each point in time. Less work up-front, more production/editing work. All easily doable, tho.

[Trevor B.]

Originally Posted by Doug Young

Well, that's not the flow I was suggesting. I suggest:

Record your "studio" audio - I assume you've already done that.

Now record one camera angle while you listen to the playback, playing along, while recording that same playback to the camera. You're listening in headphones, right? So you need a splitter that can send the headphone output to the camera's line-in. That means the camera's audio will be the original studio audio. Alternatives would be to play back the sound not over headphones, but over a speaker, loud enough that that's what the camera picks up, and you can hear it as well.

Hey Doug, while your second approach strikes me as the most efficient way to line up studio audio with video my question pertains to the first. Is the camera "line in" the same as the "mic" input on a camcorder?
Thanks.

[Doug Young]

Originally Posted by Trevor B.

Hey Doug, while your second approach strikes me as the most efficient way to line up studio audio with video my question pertains to the first. Is the camera "line in" the same as the "mic" input on a camcorder?
Thanks.

Depends on what camera you have. Some may have "line in", "mic in" or nothing at all. You need some way to get the audio into the camera. If you have a mic-in, you can probably feed that a signal from a player of some kind, you may just have to adjust the levels to not be distorting. In the worst case, a boom box playing back your guide track, sitting near the camera. The sound doesn't have to be "good", it just has to be good enough that Logic can correlate it.

[KevWind]

Originally Posted by Doug Young

Depends on what camera you have. Some may have "line in", "mic in" or nothing at all. You need some way to get the audio into the camera. If you have a mic-in, you can probably feed that a signal from a player of some kind, you may just have to adjust the levels to not be distorting. In the worst case, a boom box playing back your guide track, sitting near the camera. The sound doesn't have to be "good", it just has to be good enough that Logic can correlate it.

I'm thinking you mean "enough that FCPX can correlate it" but yes or just go back to the way he was doing it and as you stated two posts ago just dump it all into FXP and manually try to line up the action on the videos with the GB audio. By cut an paste ( and move back or forth on the timeline frame by frame if needed ) to get the video action to sync with the audio

[Doug Young]

Originally Posted by KevWind

I'm thinking you mean "enough that FCPX can correlate it" but yes or just go back to the way he was doing it and as you stated two posts ago just dump it all into FXP and manually try to line up the action on the videos with the GB audio. By cut an paste ( and move back or forth on the timeline frame by frame if needed ) to get the video action to sync with the audio

Right, sorry, FC, not Logic

[DoryDavis]

I'm in the camp of recording the track as best you can in your project studio, then shooting multiple angles of playing along with the track. I've got a cheap analog camera, and the mic on that picks up the cd playing back. Then I import the multiple takes into my software (along with the studio audio take). It is not tough to line them up using the waveform function to just see where the first note hits. Then just choose the angles during editing.
Doug's right, once you are in the play along mode, it opens the door to do a more mtv approach, with a story. Multiple angles are I guess somewhere in between, adding some visual interest, rather than a single camera shot. I posted this last week, but I'll post it again here, I shot 3 camera angles and conservatively used them each a couple times.