Welcome to our community

Be a part of something great, join today!

  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Simplifying Music videos... can Red do this?

Frank Weeks

Well-known member
Joined
Apr 19, 2007
Messages
1,483
Reaction score
0
Points
36
Location
Johnson City, TN
At the FCPUG meeting during NAB back in 08 I watched an interesting presentation by Tim Dashwood called the How to create the world's fastest music video - from shoot to edit.

You can see it here
http://www.macvideo.tv/camera-technology/features/?articleId=102467&pagType=samecat
The relevant part starts about 20 minutes in.

Briefly it explains how to use LTC timecode (sounds like a modem) on one channel of audio from an ipod or any stereo device and read by the camera to sync up multiple takes, shot from the same camera.
The other channel is sent to an audio amp for the artist to lip-sync.

My question is ...How do you input LTC from one audio channel into Red (One or Epic) where it is used as input timecode?
It is possible on vastly less capable cameras.

Would it require the Lemo input? Adapters? Any menu changes?
If the camera(s) are not capable is there a device that would perform the conversion (audio LTC to TC readable by Epic and R1)?

I would like to do the same thing that is done in the video. Shoot multiple takes from a single or multiple (Epic and Red one) if the TC can be sent and understood by both. The timecode would be identical for every note on each take. Even on partial takes, So when loaded into the multi camera feature in FCP all takes would synced instantly. Up to 16 takes. No slate required. A rough cut could be prepared in real time then tweaked to your liking. The amount of time saved in both production and post would be substantial.

I shoot music videos frequently and if this is possible it would be very helpful. For the record I asked this question a few years back without success. I'm hoping that firmware has improved and perhaps things have changed.

Thanks
 
Last edited:
Did not look trough the video link in your post but to me the sync issue is not so difficult to handle even without sync boxes etc. if you got a mic on each camera you could use a software like this one: http://www.singularsoftware.com/pluraleyes.html

There is quite a few of those programs that mach sync by analyzing the sound files. of your proxies.
 
Did not look trough the video link in your post but to me the sync issue is not so difficult to handle even without sync boxes etc. if you got a mic on each camera you could use a software like this one: http://www.singularsoftware.com/pluraleyes.html

There is quite a few of those programs that mach sync by analyzing the sound files. of your proxies.

Thanks Bjorn. I'll try out pluraleyes. Would the FCP version work in Premiere Pro too? Not critical but would be nice.

Originally Posted by Jorgen Persson

You record the LTC as audio on the Epic. Then you use a software in post to convert it to TC for your clips.
For FCP you need to use a plugin like FCP auxTC reader.
http://www.videotoolshed.com/product...p-auxtc-reader

Another good option Jorgen. This sounds very close to what I described. Thanks. Nice that both have 30 day free trials.
It doesn't appear that this can be handled by the camera(s) alone but at least it can be done.
 
Briefly it explains how to use LTC timecode (sounds like a modem) on one channel of audio from an ipod or any stereo device and read by the camera to sync up multiple takes, shot from the same camera. The other channel is sent to an audio amp for the artist to lip-sync.
I would not use an iPod for on-set playback. I don't believe they're stable enough for this -- they're going to drift quite a bit.

Short of hiring an actual Pro Tools operator on set with a referenced timecode output, I've had good luck playing back tracks with BWF-Widget Pro on a laptop. This requires that you have a Broadcast WAV file with embedded timecode, plus the track on one channel (going to PA speakers in mono) and a duplicate of the timecode on channel 2, with the song starting precisely at 01:00:00:00, and 4 beeps as a count-off prior to the song. I generally only feed the scratch track to camera and then feed the timecode out to a wireless timecode slate on set. As far as I know, this method has been done for more than 20 years and works very well.

PluralEyes will autosync to a point with the music track. I'm not sure I'd trust the timecode off the playback except for short bursts, but you figure, 99% of most music videos will involve short shots, plus editors tend to grab shots all over the place, not necessarily sync shots from that precise instant in the song.

I've worked on a couple of hundred music videos (Fleetwood Mac, Van Halen, Rolling Stones, Barbra Streisand, you name it), and the syncing process in post is not that big a deal. Literally, it's a night's work for a decent assistant editor. (Unless it's a concert, which is a different situation.)
 
Sound synch software is so incredibly efficient it's mind boggling.

Or you take the extra 5 seconds and line it up manually. I don't understand how so many people find syncing sound so frightening. (Not referring to you or anyone here, just in general)
 
Hey Jarek. There are times where synching manual is very difficult. I recently worked on a project classical indian dance, and because of time, we couldn't do the whole performance every time when changing a camera angle. I was shaking my head because to me, the music all sounded similar and couldn't match it up. Along comes the update to FCPX with multi cam editing. Low and behold, it synched up 11 performances perfectly even though all of them started and ended at various parts of the song. It literally saved me about 3 days!

Or you take the extra 5 seconds and line it up manually. I don't understand how so many people find syncing sound so frightening. (Not referring to you or anyone here, just in general)
 
There are times where synching manual is very difficult. I recently worked on a project classical indian dance, and because of time, we couldn't do the whole performance every time when changing a camera angle. I was shaking my head because to me, the music all sounded similar and couldn't match it up.

In cases like these sync to visual reference works. You can find one within a shot or waveforms on the timeline.
 
In cases like these sync to visual reference works. You can find one within a shot or waveforms on the timeline.

I have done lots of manual synching and I can sympathize with Bob. Time is often fluid in middle eastern dance. (Native American dance I am less familiar with.) It is frequently a large juggling act because you often don't have a concrete visual reference like a drum strike or hand clap in frame. You have to be very careful when you are synchronizing footfalls to the beat for example, because you may be altering the performance. Footfalls aren't always meant to happen on the beat. Some dancers, and musicians treat time as a very elastic thing; stretching and compressing it continuously in their phrasing. It can be a personally defining thing. The artist will be extremely aware and may be displeased when you alter it. Think of Nat King Cole as a prime example of this. His phrasing was unique.
 
What about plugging this timecode directly into the timecode input of the Red rather than recording as an audio track? That way all the R3D files will already be stamped with the matching playback timecode...
 
Hey Jarek. There are times where synching manual is very difficult. I recently worked on a project classical indian dance, and because of time, we couldn't do the whole performance every time when changing a camera angle. I was shaking my head because to me, the music all sounded similar and couldn't match it up.
This would be solved if you played back timecoded audio and had a visible timecode slate on set. Again, as I said earlier in this thread, I've done hundreds and hundreds of music video projects done exactly this way, going back literally to the dawn of MTV. It works fine and is still used every single day for the production of musical/dance numbers on shows like Glee and Smash.
 
Did not look trough the video link in your post but to me the sync issue is not so difficult to handle even without sync boxes etc. if you got a mic on each camera you could use a software like this one: http://www.singularsoftware.com/pluraleyes.html

There is quite a few of those programs that mach sync by analyzing the sound files. of your proxies.

Or take the extra 5 seconds and just do it manually...I've been syncing audio for years without issue.
 
the tc slates on set can read the "Audio"-TC even when the playback is at 35 or 50 fps ( when you do it right)

make sure your epic in / software Workflow can do that!!!
 
is there any reason you are not going to slave the camera to the audio track's timecode?
I mean, if you have it on hand, why not just cable in to the camera?
 
is there any reason you are not going to slave the camera to the audio track's timecode?
I mean, if you have it on hand, why not just cable in to the camera?

You can't have 25 passes of a song in editorial with identical TC. That "song" TC needs to be recorded in an auxiliary fashion (either visually on a slate or audio channel), not as the main camera timecode.

Online/offline, passing to color correction could get very messy if you have a zillion camera takes with coinciding timecode.

I do music vids a lot, and I edit my own. In 2008 Pluraleyes or built-in audio syncing wasn't really avail. Tim's method works fine if you know you have the proper plugins in post to read the audio track LTC. It's rare that you can get it to work.

I'd just use an audio tool like Pluraleyes. Or by hand. With practice, by hands and ears is the fastest.
 
I make a quicktime movie of the song with time code and a count off. I use an iPhone attached to a wireless system and an amp for playback. Every time I start a take a PA flashes the phone to the camera. Here's my iPhone in action.

http://www.youtube.com/watch?v=jFkYk3kh0HY

Then during my edit I use the same quicktime file as my first track in my NLE and only need to match one frame of the camera playback to the same timecode frame on the camera take to achieve sync. I can hear some of you saying that this is not a perfect set up, it can be a frame or two off. That's true, but regardless of how well the sync mechanics of the playback and camera are, the artist will never be in exact sync with the playback during their performance so there's always the last step of fine tuning sync during the edit.

For me it is a very cheap and flexible set-up that has worked well for a whole bunch of pieces. The last one had multiple cameras, some with scratch sound some without and 32 recorded takes over 6 locations. I took me about one hour to sync them all up.
 
Back
Top