Synching to Clapper & Camera Audio Better than Synching to Timecode w/ Double System?

[Peter Moretti]

I am trying to determine the usefulness of timecode in a double system. I have read that synching an external recorder's tracks to a clapper and the camera's audio track is not very difficult and actually more accurate than synching with timecode.

Here's an example of such advice:

"Lots of large features throughout the world are shot on 35mm cameras without TC units, and synced up the traditional way.

I've even done an HDCAM feature where we decided to go with syncing on sticks, because it made more sense for that particular project.

Also, syncing on timecode, in my experience, is not nearly as accurate as syncing on sticks. My assistant usually will adjust sync on Autosynced clips anyway."

Do you agree? Is timecode really not need if I am going to record double the exact same signal to a recorder and the camera? Doesn't the camera's audio track act as an extremely accurate timecode, making timecode redundant and not as accurate?

Or am I kidding myself and should anty up the additional four or so grand to have a timecode in/out camera and recorder?

I'm trying to decide between the XH-A1 (no timecode), XH-G1 (timecode) or HVX (firewire timecode?) as a primary camera. Audio, not image quality, looks like it will be deciding factor. The HVX has the added advantage of recording 4 tracks, uncompressed 16-bit sound.

I also need to make a decision this coming week on the Sound Devices 702 (no timecode) versus the 702T (timecode) recorder? Is the "T" version worth the extra $800?

Thanks very much for any insights.

[Brooks Harrington]

Well, it's better to have TC than not.
35mm cameras never had TC, just Key Code numbers.
Running a Nagra with crystal sync was just fine for many years.
TC is not sync, just postional reference. When using the 702T, I believe TC is a header in the file, it's not striped the whole length of the file as a reel to reel recorder would do.
I have not taken TC off of a camera to a recorder.
You can sync with camera without TC.

SMPTE and EBU are deciding on a new way to sync, or a new type of "TC"

http://www.hollywoodreporter.com/hr/...1f275a40ab8291

Whomever is doing your post editing and post sound should be consulted as to what they would like from you.

Don't forget to keep really,really complete logsheets.

http://www.locationsound.com/PDF/Sound%20Report.pdf

Somebody else help me out here.

[Seth Bloombaum]

Really hard to advise or comment without more info. Peter, what are you shooting? What style of shooting?

It's certainly true that synching on sticks can be fine, especially with something written on the sticks, slated and tailslated, voiced on the mics, reference audio on the camera for backup on sync, and where you will *always* have someone to click the sticks... and actually do it!

It can work especially well in a production that is scripted and shots tightly planned. It can be pretty challenging to ID clips without strict scene/shot/take systems on paper.

TC doesn't care about that stuff (though it can still be a really good idea for efficiency through the post process).

[Dan Keaton]

If you can afford the extra $800, I would get the "t" model.

Sound Devices timecode units are first class, it is actually a world class timecode generator bulit by Ambient and built into the Sound Devices 7 series recorders (the ones with the "t" in the model name).

This is an excellent time code generator and reader. It has all of the time code features that you would ever need (at least in my opinion). This can be the master timecode clock, it is not just a slave unit, one in which another device can be the master.

You can jam synic (set another's timecode) to your 702t's master timecode clock in under ten seconds, and then disconnect the cable, or you can leave them connected.

Once you own or use Sound Devices equipment, you will discover all of their units are first class and very well made.

In other matters, yes, the audio files are timestamped at the start of the recording, but that is all that you need. When you import your file into your editor, such as Vegas, you should be able to see the timecode at any point of you clip, if you like.

Having good timecode is a nice luxury. If you purchase it now, and later purchase a good camera with timecode (such as the XH G1, XL H1, or many others), you will be set to go.

If possible, use a clapper at the start and end of every shot. However, this is not always advisable for certain interview work, as it may make your subject nervous. This is one situation where having timecode is very nice.

Also, if you are using dual sound, then record the sound in the camera also, so that you can easily align the audio in post.

Bear in mind that, just as one of your sources said, you may need to still make some adjustments in post. There is a reason for this: the speed of sound in air versus the speed of light. If you shoot a subject, the light reaches your camera before the sound does. Depending on distance and micrphone placement, you may want to make some manual adjustments.

[Dan Bridges]

Originally Posted by Dan Keaton

Bear in mind that, just as one of your sources said, you may need to still make some adjustments in post. There is a reason for this: the speed of sound in air versus the speed of light. If you shoot a subject, the light reaches your camera before the sound does. Depending on distance and microphone placement, you may want to make some manual adjustments.

In a recent 3-cam shoot, I put two cams down on the floor angled from the L & R at a distance of about 10m from the stage. I put a 3rd cam up in the control room for wide-angle stage & front floor shots. It was situated 30m from the stage. That led to a time difference between the floor cams and the top cam of 60ms (20m differential distance, with sound travelling at 340m/s approx.) Of course I did not use the top cams soundtrack in the mix - it was too boomy/roomy. It was there only to assist with syncing, although because of the crowd noise, visual syncing turned out to be better. The sound mix came from a PA feed (Hi-MD), a stereo mic near the front of stage (MD) and both floor cam mics (one with a Rode SVM).

[Steve House]

Originally Posted by Peter Moretti

...
.
Or am I kidding myself and should anty up the additional four or so grand to have a timecode in/out camera and recorder?

I'm trying to decide between the XH-A1 (no timecode), XH-G1 (timecode) or HVX (firewire timecode?) as a primary camera. Audio, not image quality, looks like it will be deciding factor. The HVX has the added advantage of recording 4 tracks, uncompressed 16-bit sound.

I also need to make a decision this coming week on the Sound Devices 702 (no timecode) versus the 702T (timecode) recorder? Is the "T" version worth the extra $800?

Thanks very much for any insights.

Another thought is to go the route of film and couple your TC capable recorder with a smart slate that displays code jammed from the recorder and then the camera's TC own ability becomes moot. You still have the timecode reference in the video on the image of the slate itself. Align the slate's code and the NLE timeline code, drop in the audio - BWF's will align the timestamp in their header to the NLE timeline automatically - and you're in sync.

I haven't found any indication that the TC coming out of an HVX via FireWire is accessable to anything except another HVX camera slaved to it and there's no other TC I/O that I'm aware of.

[Mike Peter Reed]

"synching an external recorder's tracks to a clapper and the camera's audio track is not very difficult and actually more accurate than synching with timecode."

It's not very difficult ... but it is time consuming.

More accurate? Not likely. But I routinely see shows broadcast 2 or 3 frames out of sync. "Broadcast quality" is basically a joke when it gets to the cable provider and its MPEG boxes (or perhaps that's just NTL/Virgin Media)

As to scaring interviewees with clapperboards, a well placed snap of the fingers to camera has much the same effect providing a sync point. I would question why you were bothering to shoot an interview double system though.

[Dan Brockett]

Hi:

You can read about my experience syncing three HVX-200s with a Zaxcom DEVA for a TV pilot we shot last year. You don't need TC I/O although a T/C slate with a recorder that can generate TC isn't bad. Organization and quality sound reports are much more important than TC I/O IMHO.

http://www.kenstone.net/fcp_homepage..._brockett.html

Best of luck,

Dan Brockett

[Peter Moretti]

Originally Posted by Mike Peter Reed

...
I would question why you were bothering to shoot an interview double system though.

Mike, essentially, there is real but small chance of a theatrical release. So I want to provide sound that is of high quality and can hold up under heavy post production.

[Wayne Brissette]

I'll chime in with two things. First, I have yet to use time code. Part of that may be that most of the productions I'm working on have all gone digital and in many cases this means FireStore, which means no time code even if I wanted it. I suspect that at some point in the future time code will be implemented on the Firestores, but as it stands now, you can't jam tc to the device. That doesn't seem to bother people however, we slate each scene / interview and we're golden.

Yes, I mentioned interview... since somebody asked why on an interview you would run a double-system, read this story.

I ALWAYS run a double-system. Part of it is because I have a Deva, so I might as well use it, but part of it is experience. I have had major accounts call me freaked out because something went amiss and the audio wasn't what they were expecting on the tape. In one case, I was on a shoot for a National Geographic production and things sounded fine on-camera through the headphones. However, the best the post-production house could explain, it sounded like "phasing" was happening. The production company was of course freaked out because they thought they were going to have to fly everybody back in to do the audio. I had offered while on-location to provide them with a DVD of the audio, but they said they didn't need it. Well, the double system saved our butts and I simply shipped them a DVD with all the audio and they simply had to sync the dialog up. No insurance was required, no scheduling headaches trying to get everybody back together, nothing but some video and audio alignment issues for post.

I'm a fan of double systems regardless. You don't have to always use the backup audio, if you want you can just use the guide-track on the video, but you always have the audio somewhere else if needed.

Part of this for me is simply doing what I'm getting paid for. That is, I'm on-location to provide sound. If sound doesn't happen, or technology fails I still want to have sound. The Deva for me (and almost every other sound mixer I know) has been bulletproof. I will always get sound, I will never let a producer or director blame the sound department for the failure of a shoot. We get blamed for enough stuff as it is. ;-)

Wayne

[Glenn Chan]

1- What SMPTE and EBU are working on is timecode (or some sort of successor) to deal with things like 59.97fps shooting. Right now drop-frame TC works for 29.97... what would 59.97 timecode look like?

The other sync problem happens after production... you want to maintain A/V sync when it gets to the viewer.

2- As far as I know... some of the automated sync solutions out there are not idiot-proof and do screw up from time to time.

If you use the Ambient/Denecke boxes, they may have inaccurate sync after a long period of time (depends on temperature) or if the battery or power source runs out or is interrupted.

More info here:
http://www.trewaudio.com/PDF/wolf.pdf

Clapping sticks together is mostly idiot-proof. It is possible to have the slate out of frame or too out of focus. Though even then in a NLE it isn't that difficult to manually find some point of sync reference (e.g. when people saying plosives like b or p words; when their lips are together, there is no sound on the waveform).

[Peter Moretti]

Originally Posted by Steve House

Another thought is to go the route of film and couple your TC capable recorder with a smart slate that displays code jammed from the recorder and then the camera's TC own ability becomes moot. You still have the timecode reference in the video on the image of the slate itself. Align the slate's code and the NLE timeline code, drop in the audio - BWF's will align the timestamp in their header to the NLE timeline automatically - and you're in sync.

I haven't found any indication that the TC coming out of an HVX via FireWire is accessable to anything except another HVX camera slaved to it and there's no other TC I/O that I'm aware of.

Steve,

This looks like the route I'll go: Denecke TS-C slate and Sound Devices 702T recorder.

I'm sure I could do without it. But I seem to read about a lot of instances when trying to save money on equipment winds up cost many times more in $'s and time in post.

THANKS all for your help and suggestions; and Dan that was one heck of an account!

[Dan Brockett]

Glenn wrote..."What SMPTE and EBU are working on is timecode (or some sort of successor) to deal with things like 59.97fps shooting. Right now drop-frame TC works for 29.97... what would 59.97 timecode look like?"

Hi Glenn:

Not trying to be a nit picker, but isn't it "59.94" TC? I have never heard of "59.97" TC but it wouldn't surprise me all that much if it does exist. Jeesh, 18 ATSC HD standards, plus A modes, PSF modes, it's a confusing world with HD specs right now.

Dan

[Chuck Fadely]

I don't know about other NLE's, but in Final Cut, it's really easy to synch on the audio waveforms. So mic your camera as well as your recorder and you'll have no problem synching.

[Glenn Chan]

Dan: You're right... I'm getting all the numbers confused.